Tracking spawning migrations in the River Murray

C M Bice, B P Zampatti and W Koster

SARDI Publication No. F2019/000190-1 SARDI Research Report Series No. 1027

SARDI Aquatics Sciences PO Box 120 Henley Beach SA 5022

July 2019

Bice, C. et al. (2019) Lamprey migration in the River Murray

Tracking lamprey spawning migrations in the River Murray

C M Bice, B P Zampatti and W Koster

SARDI Publication No. F2019/000190-1 SARDI Research Report Series No. 1027

July 2019

II Bice, C. et al. (2019) Lamprey migration in the River Murray

This publication may be cited as: Bice, C. M., Zampatti, B.P. and Koster, W. (2019). Tracking lamprey spawning migrations in the River Murray. South Australian Research and Development Institute (Aquatic Sciences), Adelaide. SARDI Publication No. F2019/000190-1. SARDI Research Report Series No. 1027. 36pp.

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

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

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

© 2019 SARDI This work is copyright. Apart from any use as permitted under the Copyright Act 1968 (Cth), no part may be reproduced by any process, electronic or otherwise, without the specific written permission of the copyright owner. Neither may information be stored electronically in any form whatsoever without such permission.

SARDI Publication No. F2019/000190-1 SARDI Research Report Series No. 1027

Author(s): C. M. Bice, B. P. Zampatti and W. Koster

Reviewer(s): G. Giatas and J. Nicol

Approved by: Q. Ye Science Leader - Inland Waters and Catchment Ecology

Signed:

Date: 26 July 2019

Distribution: DEW, CEWO, MDBA, SAASC Library, Parliamentary Library, State Library and National Library

Circulation: Public Domain

III Bice, C. et al. (2019) Lamprey migration in the River Murray

TABLE OF CONTENTS

ACKNOWLEDGEMENTS ...... VI EXECUTIVE SUMMARY ...... 1 1. INTRODUCTION ...... 4 1.1. Background...... 4 1.2. Objectives ...... 5 2. METHOD ...... 7 2.1. Study area ...... 7 2.2. Acoustic and PIT tags ...... 9 2.3. Lamprey capture, tagging and release ...... 9 2.4. Lamprey tracking ...... 11 2.5. Data interpretation and analysis ...... 12 3. RESULTS ...... 13 3.1. Hydrology ...... 13 3.2. Lamprey abundance and tagging ...... 14 3.3. Post-release detections, extent and timing of migration ...... 16 3.4. Upstream rate of migration ...... 19 3.5. Fish passage and migratory delay ...... 20 4. DISCUSSION ...... 22 4.1. Tag retention and affects ...... 22 4.2. Extent, timing and rate of migration ...... 23 4.3. Fish passage ...... 24 4.4. Future research and management ...... 25 5. CONCLUSION ...... 28 6. REFERENCES ...... 29 7. APPENDIX ...... 33

IV Bice, C. et al. (2019) Lamprey migration in the River Murray

LIST OF FIGURES

Figure 1. a) Map of the River Murray from Lock 26 (Yarrawonga) downstream to the Murray Mouth showing the positioning of lock and weir structures and acoustic receivers (green circles). b) Finer resolution map of the Murray Barrages and receivers in the Lower Lakes region...... 8 Figure 2. in surgery cradle showing positioning of acoustic tag harness...... 10 Figure 3. Discharge (ML.d-1) at the South Australian border (QSA, black line) and from the Murray Barrages (blue line) from 2016–2018. Sampling periods are indicated by black bars and white dashes indicate dates on which lamprey were captured...... 14 Figure 4. Mean abundance (± SE) of pouched lamprey (fish.trap event-1) sampled from various fishways on the Murray Barrages in 2016, 2017 and 2018. GVS1 = Goolwa Barrage original large vertical-slot, GVS2 = Goolwa Barrage new large vertical-slot, MDVS = Mundoo Barrage dual vertical-slot, BCVS = Boundary Creek Barrage vertical-slot, EIDVS = Ewe Island Barrage dual vertical-slot, TVS = Tauwitchere Barrage large vertical-slot, TSVS = Tauwitchere Barrage small vertical-slot. MDVS was not sampled in 2016 and EIDVS was not sampled in 2018...... 15 Figure 5. Estimated final locations of pouched lamprey tagged with acoustic and PIT tags in 2017 (open black boxes) and 2018 (open red boxes) based on acoustic receiver and PIT reader detections. Release location in 2017 (solid black boxes) and 2018 (solid red box) are also indicated...... 17 Figure 6. Pouched lamprey entangled in a commercial fishery gill net (photo: Ricky Edgel). ....17 Figure 7. Migration timing distributions for upstream migrating pouched lamprey, presenting the median, and earliest and latest dates (error bars), of first detection on select acoustic receivers and fishways PIT readers in 2017 and 2018...... 19

V Bice, C. et al. (2019) Lamprey migration in the River Murray

ACKNOWLEDGEMENTS

Thank you to Heleena Bamford and Stuart Little (Murray-Darling Basin Authority, MDBA) for facilitating funding and managing this project, and for ongoing support of aquatic ecosystem research in the MDB. In 2018, sampling was supported by the South Australian Department of Environment and Water (DEW) under the MDBA’s The Living Murray program, while acoustic tags were supplied by the Commonwealth Environmental Water Office (CEWO). Adrienne Rumbelow, Rebecca Turner and Kirsty Wedge (DEW), and Sean Kelly and Alanna Wilkes (CEWO) are thanked for these specific contributions to the project. Thanks also to several SARDI (Arron Strawbridge, Ian McGraith and Josh Fredberg), SA Water (Michael Shelton, Ray Maynard, Bryce Buchannan, Greg Bald and Dave Bishop) and Arthur Rylah Institute (ARI) staff (David Dawson) for assistance with fieldwork. All sampling was conducted under an exemption of section 115 of the Fisheries Management Act 2007. Thanks to George Giatas, Jason Nicol and Gavin Begg (SARDI), Stuart Little (MDBA) and Sean Kelly (CEWO) for reviewing this report.

VI Bice, C. et al. (2019) Lamprey migration in the River Murray

EXECUTIVE SUMMARY Understanding of life history and spatial ecology underpins the management and conservation of anadromous fishes (i.e. species that enter rivers from the sea as adults and migrate to upstream spawning grounds, with juveniles later migrating downstream to the sea). Pouched lamprey ( australis) and short-headed lamprey ( mordax) are the only anadromous fishes native to the Murray-Darling Basin (MDB). Historical evidence suggests lamprey were once common in the River Murray with migrations potentially extending up to 2000 km upstream, but they are now rarely encountered, suggesting barriers to migration and flow regulation have impacted these species. In a contemporary context, little is known about the distribution and abundance of lamprey in the MDB, or aspects of their life history, including upstream migration. Nonetheless, in the past decade, a focus on restoring connectivity (e.g. fishway construction) and ecologically relevant components of flow regimes, under the Murray–Darling Basin Authority’s Basin Plan, has increased opportunities for migration that may aid recovery of lamprey in the Murray–Darling Basin.

The objective of the current project was to use acoustic and PIT (passive integrated transponder) telemetry to investigate the upstream spawning migrations of lamprey in the MDB over three migration seasons from 2016–2018. Specifically, the project aimed to:

1. Establish an array of acoustic receivers (VEMCO VR2W) along ~2000 km of the River Murray from the Murray Barrages to Yarrawonga, including at major tributary and anabranch junctions; 2. Collect upstream migrating lamprey from fishways on the Murray Barrages in winter– spring 2016, 2017 and 2018, and tag with PIT and/or acoustic tags; 3. Download and interpret acoustic telemetry and fishway PIT reader system data with respect to the spatio-temporal characteristics of migration, including extent, timing and rate, as well as interaction with weirs and fishways; and 4. Provide commentary on potential spawning/nursery locations for lamprey in the MDB and recommendations for future research and management.

Totals of 6 (PIT only), 53 (PIT only = 10, PIT and acoustic = 43) and 6 pouched lamprey (PIT and acoustic) were captured and tagged in 2016, 2017 and 2018, respectively. A single short-headed lamprey (PIT and acoustic) was captured and tagged in 2018. Movements of PIT tagged fish are detected by PIT reader systems when fish pass through fishways on lock and weir structures on the River Murray. Lamprey tagged in 2016 and 2017 were released immediately upstream of the 1 Bice, C. et al. (2019) Lamprey migration in the River Murray

Murray Barrages. In 2017, at least 10 acoustically tagged pouched lamprey were captured in gillnets of fishermen from the Lakes and Coorong Commercial Fishery. In 2018, to avoid further interaction with the fishery, all lamprey were transported to a release location upstream of the commercial fishery.

In 2016, no PIT tagged lamprey were detected on fishway PIT reader systems. In 2017, 21 of 44 lamprey tagged with acoustic tags were detected on the acoustic receiver array. Four PIT tagged only fish were also detected on fishway PIT reader systems. In 2018, five of six lamprey tagged with acoustic tags were detected on the acoustic array. The single short-headed lamprey tagged in October 2018 was not detected.

Across 2017 and 2018, estimated extent of migration ranged from ~20–431 km. In 2017, a proportion of acoustically tagged pouched lamprey were last detected migrating into Currency Creek (~2%) and Finniss River (~9%), tributaries of Lake Alexandrina. Nevertheless, the majority of movements were made towards or into the River Murray. In both years, several lamprey (5– 10) were last detected in the River Murray in the following reaches: Mannum–Lock 1; Lock 1–2; Lock 2–3; and Lock 3–4. The cause of cessation of migration remains unclear, but may include: location of suitable spawning sites; predation; or holding behaviour as part of staged migration. Timing of migration of pouched lamprey at the Murray Barrages extended from July to early- September, and migration continued upstream until mid-October. Based on previous studies, upstream migration may extend to January. Migration rates of individual pouched lamprey generally ranged 4–20 km.d-1, but several individuals exhibited rates of >35 km.d-1.

All tagged lamprey detected on PIT reader systems on the Lock 1, 2 and 3 fishways successfully ascended the fishways, and typically with ascent rates <4 hr. This indicates the vertical-slot fishways constructed on these weirs are effective in passing pouched lamprey. Nonetheless, some lamprey spent considerable periods of time (20 hr–29 d) immediately downstream of fishways, suggesting possible migratory delay associated with difficulty in locating fishway entrances, or that these individuals were assessing downstream areas as spawning sites.

This project has improved understanding of the upstream spawning migrations of pouched lamprey in the River Murray, particularly with regard to extent, timing and rate, as well as interaction with weirs and fishways. Upstream migration occurs over a defined period, with entry into freshwater from the sea occurring predominantly in July–August. Ensuring discharge through the barrages and fishway operation over this period is likely critical to facilitate riverine ingress. Riverine migrations potentially extend to January, and may be long-distance (100s km) and rapid.

2 Bice, C. et al. (2019) Lamprey migration in the River Murray

These migrations have been aided by fishways recently constructed under the MDBA’s Sea to Hume fishway program. The location of spawning and nursery sites for pouched lamprey may include tributaries of the Mount Lofty Ranges (e.g. Currency Creek and the Finniss River) and reaches of the lower River Murray channel. Nonetheless, identifying actual nursery sites in the lower River Murray, and potentially more broadly (e.g. the mid River Murray and tributaries) is a priority for future research. With regard to short-headed lamprey, details on contemporary migrations and distribution remain unknown.

Keywords: diadromous, Geotria, Mordacia, fishway, acoustic telemetry, passive integrated transponder.

3 Bice, C. et al. (2019) Lamprey migration in the River Murray

1. INTRODUCTION

1.1. Background

Understanding of life history and spatial ecology, including patterns of movement and habitat use, underpins the management and conservation of fishes (Cooke et al. 2016). This is particularly true for diadromous fishes, which migrate between marine and freshwater habitats to complete essential life history processes (e.g. spawning, growth and maturation) (McDowall 1988). Anadromous fishes, such as lamprey, enter rivers from the sea as adults and migrate to upstream spawning grounds, with juveniles later migrating downstream to the sea. These migrations render lamprey susceptible to human modification of river systems (Jonsson et al. 1999). Specifically, flow regulating structures (e.g. dams, weirs, barrages) may delay or obstruct upstream migration, whilst alteration to flow regimes (e.g. reduced magnitude, altered timing) may disrupt migratory cues and degrade habitat quality (Quinn and Adams 1996). As such, in regulated rivers worldwide, a key objective of management is the promotion of connectivity and migration to support the population dynamics of anadromous fish over spatial scales relevant to life histories and population dynamics.

In the Murray–Darling Basin (MDB), ’s longest and most economically important river system (ABS/ABARES/BRS 2009), pouched lamprey (Geotria australis) and short-headed lamprey () are the only native anadromous fishes. Both species exhibit a life history characteristic of all anadromous lamprey, including a parasitic marine phase, and upstream adult migrations to freshwater spawning and nursery habitats (Moser et al. 2015). In several species, physico-chemical (e.g. salinity) and olfactory cues of riverine origin (e.g. pheromones from juveniles) may stimulate and guide upstream migration (Meckley et al. 2014, Bett and Hinch 2016). Juveniles (ammocoetes) reside in freshwater for several years before metamorphosing into sub-adults (macropthalmia) and migrating downstream to marine habitats. Globally, this life history has predisposed anadromous lampreys to the impacts of river regulation, and resulted in population declines in many species (Moser et al. 2015).

Investigations on the biology and ecology of pouched lamprey have largely been limited to rivers of south-western Australia and (e.g. Potter et al. 1980, Jellyman et al. 2002, Baker et al. 2017), whilst research on short-headed lamprey is even sparser (Potter 1970). Indeed, little is known regarding both lamprey species in the MDB, including the extent of migration, location of spawning/nursery habitats and population dynamics. Anecdotally, lamprey in the MDB were once considered common, with migrations potentially extending up to 2000 km upstream, but they

4 Bice, C. et al. (2019) Lamprey migration in the River Murray are now rarely encountered, potentially due to the impacts of flow regulation and barriers to migration (Potter and Strahan 1968, Maitland et al. 2015, Bice et al. 2018).

Recent focus on restoring connectivity and flow regimes under the Basin Plan (MDBA 2009) have the potential to support rehabilitation of lamprey populations in the MDB. From 2001–2013, an ambitious program was undertaken to reinstate connectivity and opportunities for along 2000 km of the main-stem of the River Murray from the sea to Hume Dam, through the construction of fishways on the Murray Barrages and 13 main channel weirs (Barrett and Mallen- Cooper 2006). This program has potentially restored access to upstream spawning habitats from which lamprey were largely excluded since the construction of the main channel weirs in the 1920s. In addition, a considerable volume of environmental water has now been recovered under the Basin Plan (~2000 GL.yr-1), with the delivery of this water guided by state priorities and the Basin-wide environmental watering strategy (MDBA 2014). Explicit expected outcomes within the strategy include improving end-of-system flow and connectivity through the Murray Barrages and Murray Mouth, increased rates of fish movement and expanded distributions, and specifically, increased movement of short-headed lamprey and pouched lamprey through key fish passages.

The first investigation of pouched lamprey movement in the MDB occurred in 2015, when 55 pouched lamprey captured at the Murray Barrages were implanted with PIT (passive integrated transponders) tags and released upstream. Subsequent detections on fishway PIT reader systems suggested upstream migrations, supported by the fishways, extend at least as far as Lock 11 (Mildura, 878 km from the Murray Mouth) (Bice et al. 2018). Whilst PIT telemetry is useful in tracking these gross movements in the River Murray, and assessing intra-fishway behaviour, it has limited capacity to investigate questions at finer spatial and temporal scales, including movement into tributaries, the location of spawning habitats, and migratory delay below barriers. Acoustic telemetry, however, provides a means of investigating such questions.

1.2. Objectives

The objective of the current project was to use both acoustic and PIT telemetry to investigate the upstream spawning migrations of adult lamprey in the MDB. Specifically, the project aimed to:

1. Establish an array of acoustic receivers along the River Murray from the Murray Barrages to Yarrawonga (~2000 river km), including at major tributary and anabranch junctions; 2. Collect upstream migrating lamprey from fishways on the Murray Barrages in winter– spring 2016, 2017 and 2018, and tag with PIT and/or acoustic tags;

5 Bice, C. et al. (2019) Lamprey migration in the River Murray

3. Download and interpret PIT/acoustic telemetry data with respect to the spatio-temporal characteristics of migration, including extent, timing and rate, as well as interaction with weirs and fishways; and 4. Provide commentary on potential spawning/nursery locations for lamprey in the MDB and recommendations for future research and management.

6 Bice, C. et al. (2019) Lamprey migration in the River Murray

2. METHOD

2.1. Study area

The Murray–Darling Basin (MDB) covers an area of >1 million km2. The southern part of the basin, comprising the River Murray and its tributaries, typically contributes the bulk of flow, and is extensively regulated by dams and weirs as well as the Murray Barrages near the river mouth (Walker and Thoms 1993) (Figure 1). The lower River Murray is unique due to a lack of major tributaries downstream of the Darling River junction (a distance of approximately 760 km). The River Murray discharges into Lake Alexandrina (>750 km2), before reaching the Coorong Estuary and Southern Ocean. Several small tributaries (Finniss River, Currency Creek, Bremer River, Angas River) discharge into Lake Alexandrina, which is disconected from the river estuary (the Coorong) by the Murray Barrages. Overall end-of-system discharge has been greatly reduced due to river regulation and water abstraction with mean discharge (4723 GL.yr-1) now approximately a third of natural (12,233 GL.yr-1).

All main channel weirs on the River Murray, and the Murray Barrages, have been retrofitted with fishways to facilitate upstream fish movement (Baumgartner et al. 2014). At the Murray Barrages, a total of ten fishways had been constructed at the time of the study, comprising varying vertical- slot and rock ramp designs (Bice et al. 2017a). The main channel River Murray weirs comprise varying designs: 1) 1:32 slope vertical-slot fishways (Locks 1 and 7–10); 2) 1:23 slope vertical- slot fishways paired with a fishlock (Locks 2–6), 3) a 1:18 slope vertical-slot fishway (Lock 26); or 4) Denil fishways (Lock 11 and 15).

7 Bice, C. et al. (2019) Lamprey migration in the River Murray a)

b)

Pomanda Point

GB MB

BCB EB TB

Figure 1. a) Map of the River Murray from Lock 26 (Yarrawonga) downstream to the Murray Mouth showing the positioning of lock and weir structures and acoustic receivers (green circles). b) Finer resolution map of the Murray Barrages (GB = Goolwa Barrage, MB = Mundoo Barrage, BCB = Boundary Creek Barrage, EB = Ewe Island Barrage and TB = Tauwitchere Barrage) and receivers in the Lower Lakes region.

8 Bice, C. et al. (2019) Lamprey migration in the River Murray

2.2. Acoustic and PIT tags

Lamprey were tagged with a combination of acoustic and PIT (passive integrated transponder) tags, or PIT tags only. Acoustic tags were either VEMCO V7-4L (dimensions 7 x 22.5 mm, 1.0 g dry weight) or V9-2L (9 x 29 mm, 2.9 g) transmitters (VEMCO, Halifax, NS, Canada) with an average delay between transmission of 120 s and projected battery lives of ~305 (V7) and ~550 days. Tag size used aimed to maintain a transmitter to fish weight ratio of <2% (Jepsen et al. 2002).

PIT tags were Texas Instruments RI-TRP-REHP half-duplex eco-line glass transponders (3.85 x 23.1 mm, 0.6 g). PIT tags, unlike acoustic tags, are not limited by battery life, and provide a complementary means of tracking fish movement, particularly when passage past barriers is facilitated by fishways with associated PIT reader systems.

2.3. Lamprey capture, tagging and release

Pouched lamprey were sampled from fishways on the Murray Barrages using fishway traps (see Bice et al. 2016, 2017). Sampling was conducted from 12 July–31 August 2016 (n = 9–12 sampling events per fishway), 4 July–15 September 2017 (n = 10–24) and 10 July–5 October 2018 (n = 7–13) at vertical-slot fishways on Goolwa, Mundoo, Boundary Creek, Ewe Island and Tauwitchere barrages. Sampling was specifically extended into October in 2018 in an effort to sample short-headed lamprey, which likely migrate later into spring than pouched lamprey (Bice et al. 2018). During each sampling event, traps were set overnight, and deployed and retrieved using a crane truck. Upon retrieval, all trapped fish were removed and placed into aerated holding tanks. Lamprey were sorted from the catch for processing and all remaining fish released upstream.

In 2016, efforts were made to insert acoustic tags in the peritoneal cavity of lamprey, but this approach was unsuccessful due to limited space in the cavity. Thus, in 2016, lamprey were tagged with PIT tags only, and in 2017 and 2018, an external tagging method was subsequently developed for acoustic tagging.

Following capture, lamprey were anaesthetised individually using a 0.05 ml.L-1solution of AQUI- S (AQUI-S, Lower Hutt, New Zealand). Anaesthesia times were typically <10 min. Individuals were then measured for total length (mm) and weight (g), and placed in an elongated v-shaped support. For PIT tagging, a short (~5 mm) incision was made on the ventral surface, anterior to the first dorsal fin, and the PIT tag inserted into the peritoneal cavity. Surgical openings were

9 Bice, C. et al. (2019) Lamprey migration in the River Murray closed with Vetbond surgical glue (3M, Sydney, NSW, Australia) and procedures typically completed within one minute.

The external attachment method developed for acoustic tagging was adapted from techniques previously used on flatfishes (e.g. winter flounder (Pseudopleuronectes americanus); DeCelles and Cadrin 2010). Acoustic tags were secured in a ~40 mm long harness of soft latex tubing (7 mm internal diameter for V7 tags, and 9 mm internal diameter for V9 tags; Gecko Optical, Sterling, WA, Australia) with two-part epoxy. Nickel pins were then passed through a Peterson disc, the latex harness and then the dorsal musculature beginning ~50 mm anterior the first dorsal fin and ~5 mm below the dorsal midline. The nickel pins were then fixed on the opposite side of the lamprey by means of a soft latex tab, Peterson disc and plastic earring backings (Figure 2). Surgical procedures typically took <6 min. A laboratory holding trial was undertaken during the development of the technique using the morphologically similar short-finned (Anguilla australis) and yielded a tag retention rate of ~85% over a period of five months.

Figure 2. Pouched lamprey in surgery cradle showing positioning of acoustic tag harness.

10 Bice, C. et al. (2019) Lamprey migration in the River Murray

Prior to release, all tagged lamprey were observed until they exhibited natural behaviors (e.g. ability to maintain balance and swim freely), which typically occurred within 10 min of procedures. In 2016 and 2017, tagged lamprey were released immediately upstream of the barrage of their capture, in a manner that minimised risk of ‘fallback’ through the barrages. In 2018, to minimise capture of tagged lamprey in gillnets used by fishermen from the Lakes and Coorong commercial fishery, tagged lamprey were transported by car in an aerated holding container and released upstream of the Lower Lakes, at Tailem Bend (~86 km MTD, middle thread distance).

2.4. Lamprey tracking

In 2016, to monitor movements of acoustically tagged pouched lamprey, an array of 47 receivers (VEMCO VR2W model) was established in the River Murray from the Murray Mouth to Yarrawonga (a distance of ~2000 km) (Figure 1). Receivers were placed: 1) at key tributary and anabranch junctions to determine direction of movement; 2) below weirs to investigate potential obstruction of upstream migration; and 3) between distantly spaced junctions and regulatory structures to provide greater resolution within long river reaches. In 2018, receivers in the lower reaches of Eastern Mount Lofty Ranges tributaries (i.e. the Finniss, Bremer and Angas rivers, and Currency Creek) were removed given that tagged lamprey were released 60–80 km upstream. Receivers were downloaded on four occasions, approximately every four months, from November 2017 to February 2019.

All fishways on main channel weirs of the River Murray from Lock 1 to Lock 26 are fitted with KarlTek KLK5000 PIT reader systems (KarlTek Pty Ltd, Melbourne, Australia). These systems are half- and full-duplex compatible and comprise multiple digital-signal-processing (DSP) decoders (‘readers’) and water proof antennas. The antennas are mounted around the vertical- slots of each fishway to detect tagged fish passing through. The number of antennas at each fishway ranges from two to five; at a minimum, each fishway has an antenna at the fishway entrance and exit slots, while some fishways have additional antennas between the entrances and exits. The systems are auto-tuning, providing optimal tag read-range year-round, and data is telemeted to a central database. At sites where vertical-slot fishways are paired with fishlocks (Locks 2–6), PIT readers are not installed on fishlocks. The MDB fishway PIT telemetry database was interrogated for detections of pouched lamprey tagged in the current study.

11 Bice, C. et al. (2019) Lamprey migration in the River Murray

2.5. Data interpretation and analysis

Migration of lamprey was characterised by calculating the extent, timing and rate of movement. Extent of migration was defined as the reach in which migration appeared to cease based upon detections on acoustic receivers and PIT reader systems. Migration timing was estimated for the point of capture (Murray barrages = 7 km MTD), and select acoustic receivers (Clayton = 23.5 km MTD, Wellington = 74 km MTD, Mannum = 150 km MTD, DS Lock 1 = 274 km MTD, DS Lock 2 = 362 km MTD and DS Lock 3 = 431 km MTD) and PIT system locations (Lock 1–3), and presented as median dates, as well as date of first and last detections at each site. Migration rates (km.d-1) were calculated from date–time stamps from acoustic receivers and PIT reader systems. For PIT tagged only fish, these rates were calculated for the reach between the release locations (Murray Barrages) and Lock 1, and subsequent reaches between fishway PIT reader systems. For fish acoustically tagged in 2017, rates were calculated between release location and Wellington, and Wellington and Mannum. For fish acoustically tagged in 2018, rates were calculated between release location (Tailem Bend) and Mannum, and Mannum and Lock 1. For all reaches, migration rates were calculated as time elapsed between last detection on a downstream receiver/system and first detection on the next upstream receiver/system, and are presented as medians as the data was not normally distributed.

Passage through fishways was assessed by interrogating detection records from PIT reader systems and calculating passage efficiency, the number of attempts made by each individual and ascent rates for individuals that successfully passed the fishways. Passage efficiency is the proportion of tagged lamprey detected within the fishways that successfully ascended the fishways. Fishway entrance antennas detect the presence of individuals, but not whether an individual is ascending or descending, and thus interval analysis was used to differentiate ascent attempts. A time interval of 2 hrs between consecutive detections at entrance antennas was adopted to differentiate ascent attempts and was based on the minimum time required by pouched lamprey to ascend the longest 1:23 fishways on the River Murray in 2015 (Bice et al. 2018). Two proxy metrics – time to pass and time spent downstream – were then used as indicators of potential migratory delay. Time to pass was calculated as the time elapsed between first detection on a fishway entrance and last detection on the fishway exit antenna, and incorporates multiple attempts and ascents if undertaken, and reflects the true impediment of passage through the fishway. Time spent downstream was calculated as the time elapsed between first detection on an acoustic receiver downstream of a weir and either, first detection on the associated fishway entrance, or cessation of detections on the acoustic receiver downstream.

12 Bice, C. et al. (2019) Lamprey migration in the River Murray

3. RESULTS

3.1. Hydrology

From 2016–2018, discharge in the lower River Murray at the South Australian border (QSA), was generally characterised by low within-channel flows (<10,000 ML.d-1), punctuated by an overbank flood that peaked at ~94,000 ML.d-1 in November 2016, and a smaller within-channel flow pulse of ~17,000 ML.d-1 in December 2017 (Figure 3). Barrage discharge followed a similar pattern, albeit attenuated relative to QSA. In 2016, fish sampling at the Murray Barrages occurred in association with relatively high cumulative barrage discharge (mean ± SD = 16,228 ± 14,516 ML.d-1), but with substantial daily variability due to storm events, reverse flows and subsequent periods of barrage closure (i.e. discharge range 0–48,353 ML.d-1). In 2017, moderate discharge occurred during sampling, ranging from 0–14,757 ML.d-1, with a mean of 4773 ± 3994 ML.d-1, whilst in 2018, sampling occurred in association with lower discharge (range = 0–10,733 ML.d-1; mean ± SD = 1575 ± 1899 ML.d-1). Barrage discharge during sampling in all three years was supported by provision of environmental water allocations from the Commonwealth Environmental Water Holder, and fishways on the barrages operated continuously.

13 Bice, C. et al. (2019) Lamprey migration in the River Murray

120000

100000

)

-1 80000 QSA Barrages

60000

40000

Discharge (ML.d Discharge

20000

0 Jan Jul Jan Jul Jan Jul Jan

2016 2017 2018

Figure 3. Discharge (ML.d-1) at the South Australian border (QSA, black line) and from the Murray Barrages (blue line) from 2016–2018. Sampling periods are indicated by black bars and white dashes indicate dates on which lamprey were captured.

3.2. Lamprey abundance and tagging

Six, 53 and six pouched lamprey were captured in 2016, 2017 and 2018, respectively. Lamprey were predominantly collected at Goolwa Barrage in 2016, with an individual also collected at Boundary Creek (Figure 4). In 2017, pouched lamprey were sampled from five fishways across Goolwa, Mundoo and Tauwitchere barrages, and were most abundant at Mundoo and Tauwitchere. In 2018, pouched lamprey were again most abundant at Goolwa, with individual lamprey also sampled from Mundoo and Tauwitchere. In October 2018, during fishway monitoring as part of The Living Murray Program, a single short-headed lamprey was sampled from Goolwa Barrage.

14 Bice, C. et al. (2019) Lamprey migration in the River Murray

1.6

1.4 GVS1 GVS2 MDVS 1.2 BCVS EIDVS ± SE TVS

-1 1.0 TSVS

0.8

0.6

0.4

lamprey . trap event

0.2

0.0 2016 2017 2018

Figure 4. Mean abundance (± SE) of pouched lamprey (fish.trap event-1) sampled from various fishways on the Murray Barrages in 2016, 2017 and 2018. GVS1 = Goolwa Barrage original large vertical-slot, GVS2 = Goolwa Barrage new large vertical-slot, MDVS = Mundoo Barrage dual vertical-slot, BCVS = Boundary Creek Barrage vertical-slot, EIDVS = Ewe Island Barrage dual vertical-slot, TVS = Tauwitchere Barrage large vertical-slot, TSVS = Tauwitchere Barrage small vertical-slot. MDVS was not sampled in 2016 and EIDVS was not sampled in 2018.

In 2016, all lamprey captured (mean TL ± SE = 554 ± 11 mm) were tagged with PIT tags only (Appendix 1). In 2017, 43 lamprey were tagged with a combination of acoustic and PIT tags (mean TL ± SE = 560 ± 5 mm, mean weight ± SE = 166 ± 4 g), while a further 10 lamprey were tagged with PIT tags only (mean TL ± SE = 532 ± 10 mm, mean weight ± SE = 166 ± 4 g) (Appendix 1). For eight of these ten lamprey, acoustic tags were not attached due to injuries sustained by individuals from potential predation events prior to capture. In 2018, all pouched lamprey (mean TL ± SE = 574 ± 10 mm, mean weight ± SE = 169 ± 11 g) and the single short-headed lamprey (TL = 411 mm, weight = 70 g) captured were tagged with a combination of acoustic and PIT tags (Appendix 1).

15 Bice, C. et al. (2019) Lamprey migration in the River Murray

3.3. Post-release detections, extent and timing of migration

In 2016, no PIT tagged lamprey were detected on PIT reader systems following release. In 2017, 21 of 44 (48%) lamprey tagged with acoustic tags were subsequently detected on the acoustic receiver array, with one also detected on the Lock 1 PIT reader system (Appendix 1). Four PIT tagged only lamprey were also detected on PIT reader systems (Locks 1, 2 and 3). In 2018, five of six lamprey tagged with acoustic tags were detected on the acoustic array, with two also detected on PIT reader systems (Locks 1 and 3). The single short-headed lamprey tagged in October 2018 was not detected.

In 2017, estimated extent of migration of pouched lamprey included Currency Creek, the Finniss River, Goolwa Channel, Lake Alexandrina, and several reaches of the River Murray including downstream Lock 1, Lock 1–2, Lock 2–3 and Lock 3–4 (Figure 5). Lamprey were tagged across three barrages and the number of subsequent detections, and direction of movement (if discernible), varied among tagging locations (Appendix 1). Of the 10 lamprey tagged and released at Goolwa Barrage, four moved east into Lake Alexandrina, being last detected on the receiver at Clayton, three were last detected moving into the Finniss River, one into Currency Creek, one in the Goolwa Channel and one was not detected.

Of 19 fish tagged and released at Mundoo Barrage, three were last detected at Clayton, one in Goolwa Channel and two crossed Lake Alexandrina and were detected at Wellington (74 km MTD), whilst 13 were not detected. Of 14 fish tagged and released at Tauwitchere Barrage, two were last detected at Clayton with one moving into the Finniss River, whilst four individuals crossed Lake Alexandrina and were last detected at either Wellington, Mannum (150 km MTD) or Lock 1 (274 km MTD). Seven individuals were not detected. In total, six of 43 acoustically tagged lamprey (14%) were detected in the River Murray at Wellington, Mannum or Lock 1, with one individual passing Lock 1. In contrast, four of ten pouched lamprey (40%) that were PIT tagged only were subsequently detected ascending the Lock 1 fishway, with two passing the Lock 2 fishway (362 km MTD), and a single individual passing the Lock 3 fishway (431 km MTD).

Reports from a commercial fishermen indicated that at least 10 acoustically tagged lamprey had been incidentally captured in gill-nets in the Lakes and Coorong commercial fishery, as a result of the external tagging technique used (Ricky Edgel Pers. Comm.; Figure 6), which caused mortality in most instances. Interaction with the commercial fishery was likely the cause of low numbers of acoustically tagged lamprey being detected upstream in the River Murray and an estimated 29 individuals ceasing migration in Lake Alexandrina.

16 Bice, C. et al. (2019) Lamprey migration in the River Murray

3 1 1 1 1 2 2

3 1 Finniss River 4 29 1 2 1

Goolwa Channel

Figure 5. Estimated final locations of pouched lamprey tagged with acoustic and PIT tags in 2017 (open black boxes) and 2018 (open red boxes) based on acoustic receiver and PIT reader detections. Release location in 2017 (solid black boxes) and 2018 (solid red box) are also indicated.

Figure 6. Pouched lamprey entangled in a commercial fishery gill net (photo: Ricky Edgel).

17 Bice, C. et al. (2019) Lamprey migration in the River Murray

In 2018, transporting pouched lamprey to a release location upstream of the fishery was associated with improved detection rate. Estimated extent of migration included the Goolwa Channel and three reaches of the River Murray: 1) downstream Lock 1; 2) Lock 2–3; and 3) Lock 3–4. Of six individuals tagged, four were detected on receivers or fishways upstream of the release location, with two being last detected on acoustic receivers immediately downstream of Lock 1 and one immediately downstream of Lock 3. Another individual was last recorded on an acoustic receiver at Mannum, but was subsequently detected on PIT readers passing through fishways on Lock 1 and Lock 3, suggesting it had shed its acoustic tag. Both individuals that were detected on the acoustic receiver downstream of Lock 3 were not detected on the acoustic receiver downstream of Lock 2 or passing through the Lock 2 fishway. One individual moved downstream from the Tailem Bend release location, being subsequently detected on the receivers at Wellington and then Clayton.

Detections of pouched lamprey were recorded between July and October in 2017, and July and September 2018 (Figure 7). In 2017, detections on receivers <30 km upstream of the barrages occurred from early July–late September, reflecting the broad period over which lamprey were tagged in 2017. In 2017 and 2018, as expected, transit through upstream reaches (i.e. Wellington, Locks 1, 2 and 3) occurred progressively later in the year than downstream reaches. For instance, first detections of individual lamprey at Wellington occurred in early August 2017 and late July 2018, with first detections at Lock 3 occurring in October 2017 and September 2018.

18 Bice, C. et al. (2019) Lamprey migration in the River Murray

500

2017 2018 400

300

200

River km (MTD) River 100

0

Jul Aug Sep Oct Nov

Figure 7. Migration timing distributions for upstream migrating pouched lamprey, presenting the median, and earliest and latest dates (error bars), of first detection on select acoustic receivers and fishways PIT readers in 2017 and 2018.

3.4. Upstream rate of migration

Differing technologies (i.e. acoustic and PIT) used to track lamprey movements and different release points among years, precludes quantitative comparison of migration rates. Nonetheless, 25 estimated migration rates were calculated for seven reaches: 1) barrages to Lock 1; 2) Barrage to Wellington; 3) Wellington to Mannum; 4) Tailem Bend to Mannum; 5) Mannum to Lock 1; 6) Lock 1 to Lock 2; and 7) Lock 2 to Lock 3 (Figure 8). Rates varied among individuals and reaches, but ranged over similar values in 2017 (3.6–44 km.d-1) and 2018 (4.6–41.8 km.d-1). For lamprey with acoustic tags in 2017, migration rates were slowest between the Murray Barrages and Wellington (median = 5.9 km.d-1) and increased between Wellington and Mannum (median = 10.4 km.d-1). This pattern concurs with the median migration rate of PIT tagged only fish between the barrages and Lock 1 (9.6 km.d-1), which was intermediate between the two. Migration rates for PIT tagged only fish between Lock 1 and Lock 2 (median = 26 km.d-1), and between Lock 2 and 19 Bice, C. et al. (2019) Lamprey migration in the River Murray

Lock 3 (37.5 km.d-1) were more rapid. In 2018, median migration rate from Tailem Bend to Mannum was 24 km.d-1 and from Mannum to Lock 1 was 11.3 km.d-1.

50

2017 40 2018

)

-1

30

20

Migration rate rate (km.d Migration

10

0

Lk 1 to Lk 2 Lk 2 to Lk 3

Barrage to Lk1

Mannum to Lk 1

Tailem to Mannum

Barrage to Wellington Wellington to Mannum

Figure 8. Median (dot) and range (error bar) of migration rates (km.d-1) of pouched lamprey in seven reaches of the lower River Murray across 2017 and 2018. Reaches are progressively further upstream from left to right.

3.5. Fish passage and migratory delay

In 2017, five pouched lamprey were detected on the Lock 1 fishway PIT reader, two at Lock 2, and one at Lock 3. All individuals that were detected on a given fishway, successfully ascended the fishway (i.e. 100% passage efficiency). All lamprey ascended the fishways in a single attempt, except for one individual, that passed Lock 1, 2 and 3. This lamprey made 2–3 successful ascents at each fishway suggesting that some ascents were followed by weir fallback (i.e. passing downstream over the weir crest following ascent of the fishway). Ascent rates ranged 57 min to 1 hr 56 min at Lock 1 (n = 7), 2 hr 20 min to ~21 hr at Lock 2 (n = 3) and 2 hr 21 min to ~4 d at Lock

20 Bice, C. et al. (2019) Lamprey migration in the River Murray

3 (n = 3). High variability in time to pass was due to one individual that undertook multiple ascents that ranged 57 min to 5 d at Lock 1, 2 hr 20 min to 2 d at Lock 2, and 5 d at Lock 3. Only one of the lamprey detected at Lock 1 had originally been acoustic tagged, but it had likely shed its acoustic tag, and thus, time spent in the vicinity downstream of the structure could not be calculated.

In 2018, two lamprey were detected on the Lock 1 fishway PIT reader, none at Lock 2, and one at Lock 3. Non-detection at Lock 2, suggests the individual that reached Lock 3 likely passed through the navigation lock at Lock 2. Both individuals detected on fishways, successfully ascended (i.e. 100% passage efficiency). Both lamprey that ascended Lock 1, did so in single attempts, whilst the lamprey that ascended Lock 3, did so on four occasions, suggesting some ascents were followed by weir fallback. Ascent rates at Lock 1 ranged 2 hr 16 min to 3 hr 7 min (n = 2), and at Lock 3 ranged from 3 hr 36 min to 1 d 8 hr (n = 3). For the lamprey that passed Lock 3, time to pass was ~7 d given multiple successful attempts and weir fallback. This individual had likely shed its acoustic tag, and thus, time spent downstream could not be calculated for Lock 1 or Lock 3. Based on acoustic receiver detections of the other individual that passed the Lock 1 fishway, it spent ~20 d in the downstream vicinity of Lock 1 before locating and passing through the fishway. This same individual was also detected on the acoustic receiver immediately downstream of Lock 3, where it was detected intermittently over 29 d, but it did not enter the fishway. In addition, two further acoustic tagged lamprey were detected immediately downstream of Lock 1 for durations of 20 hr and 9 d, respectively, but did not enter the fishway.

21 Bice, C. et al. (2019) Lamprey migration in the River Murray

4. DISCUSSION The current study represents the first targeted investigation of upstream migration of pouched lamprey in the River Murray. The project has generated substantial new knowledge on the species’ ecology in the MDB and in the southern hemisphere. Specifically, the extent, timing and rate of migration, interaction with weirs and fishways, and potential multiple spawning/nursery locations for lamprey in the southern MDB. The results are discussed in the context of previous research on pouched lamprey and other lamprey species, with recommendations for future research and management.

4.1. Tag retention and affects

In 2017, mortality of acoustically tagged lamprey due to capture in fishing nets was substantial. The main channel of the lower River Murray, between the Wellington Ferry and Pomanda Point, is an area of high fishing pressure, where acoustically tagged lamprey readily became entangled in gill-nets, likely due to the external tag attachment. Whilst 10 interactions with gill-nets were reported for acoustically tagged fish, other interactions may have occurred that were not observed or reported. For comparison, 40% of individuals that were PIT tagged only and released at the Murray Barrages were detected on the Lock 1 PIT reader, but just 2.3% of acoustically tagged lamprey were detected.

In 2018, releasing acoustically tagged lamprey at Tailem Bend, upstream of the commercial fishery resulted in greater overall detection rates on the acoustic array (2017 = 43%, 2018 = 83%), and tag shedding was confirmed for only one individual. The remaining lamprey were detected over linear ranges of 63–345 km and periods of up to 95 d, with ~66% (n = 4) reaching the vicinity of Lock 1, while the proportion of individuals detected on the Lock 1 PIT system (~33%, n = 2) was comparable to PIT tagged only fish in 2017. While the external attachment method presented greater risk of tag fouling and shedding than internal implantation, data from 2018 suggest that when greater numbers of lamprey are tagged, and individuals are released upstream of the commercial fishery, retention rates and periods are likely sufficient to generate data adequate to investigate the movements of pouched lamprey.

Sample sizes to determine migration rates for acoustic tagged lamprey were generally low, and calculated over different river reaches to PIT tagged only fish. Nevertheless, our results suggest that acoustic tagging did not overly impact migration rate. In 2018, median upstream migration rates of acoustically tagged fish between the Tailem Bend release site and receivers at Mannum and downstream Lock 1 were 11 and 24 km.d-1, respectively. In comparison, in 2017, the median

22 Bice, C. et al. (2019) Lamprey migration in the River Murray migration rate between the barrages and Lock 1 for PIT tagged only fish was 9.6 km.d-1, and in a previous study in 2015 was 15.7 km.d-1 (SARDI unpublished data).

4.2. Extent, timing and rate of migration

No lamprey that were PIT tagged and released in 2016 were subsequently detected on fishway PIT readers. This may be a result of individuals selecting to migrate upstream into local tributaries (e.g. the Finniss River) rather than the River Murray, or due to the high flow conditions experienced in the River Murray in 2016. From August to December 2016, flow at the South Australian border ranged 24,000–94,000 ML.d-1; during high flows, weirs 1–11 are ‘removed’ as water levels rise, and the discharge at which this occurs ranges 29,000–68,000 ML.d-1 (Bice et al. 2017b). Under such conditions, lamprey may pass weir structures directly, rather than being restricted to fishways.

Across 2017 and 2018, estimated extent of migration of acoustically and PIT tagged pouched lamprey ranged from ~20–431 km MTD. In 2017, movements were predominantly made into Lake Alexandrina and toward the River Murray, but a proportion of acoustically tagged pouched lamprey were last detected migrating upstream into Currency Creek (~2%) and Finniss River (~9%). This represents the first verified record of a pouched lamprey from Currency Creek, and just the second from the Finniss River (2001; Hammer et al. 2009). In 2017 and 2018, several lamprey were last detected in the River Murray in the following reaches: Mannum–Lock 1; Lock 1–2; Lock 2–3; and Lock 3–4. In 2015, in the same reaches ~65% of PIT tagged lamprey were estimated to have ceased migration, while the remaining lamprey ceased migration in reaches between Lock 6 and Lock 15 (Bice et al. 2018). The cause of cessation of migration in particular reaches is unclear, but may include: the location of favourable spawning sites; predation; or holding behaviour as part of staged migration. Following initial freshwater entry and upstream migration, northern hemisphere lamprey (e.g. , tridentatus) often exhibit an ‘over-wintering’ holding phase before recommencing movement immediately prior to spawning in spring (Clemens et al. 2010). Recommencement of migration was not observed in this study, nor during the 2015 study in the River Murray (Bice et al. 2018), or in previous studies of pouched lamprey movement from New Zealand (Kelso and Glova 1993, Jellyman et al. 2002). Thus, we suggest cessation of migration is likely related to spawning or predation, but is an important area for future research.

23 Bice, C. et al. (2019) Lamprey migration in the River Murray

In this study, the migration timing for pouched lamprey extended from July to mid-October, and as would be expected, migration timing was progressively later with increasing distance upstream. This result is consistent with migration timing to at least Lock 6 (620 km MTD) for lamprey PIT tagged in 2015. Nevertheless, the temporal extent of upstream migration likely extends over a broader period than indicated by data from 2017 and 2018 alone. In 2015, a PIT tagged lamprey was detected at Lock 10 (Wentworth, 825 km MTD) in early December, whilst newspaper articles in 1949/50 note presence of ‘lamprey’ at Mildura (Lock 11) and Euston (Lock 15) in January (The Riverine Herald 1950). Consequently, run timing for pouched lamprey in the southern MDB likely extends from at least July to January.

Migration rates of individual pouched lamprey generally ranged 4–20 km.d-1, but several individuals exhibited rapid rates of >35 km.d-1. These rates of movement, and overall distances moved, are greater than reported by two other studies that used radio-telemetry to investigate movement of lamprey in New Zealand streams (Kelso and Glova 1993, Jellyman et al. 2002). In those studies, overall distances were generally <20 km with rates of typically <0.5 km.d-1. Such discrepancy is likely due to the River Murray being far greater in length and discharge than the New Zealand streams, and the presence of suitable spawning sites short distances upstream in those streams.

Migration rates in the current study also were generally lowest in reaches between release locations (i.e. the barrages and Tailem Bend) and next detection, and greatest in the riverine reaches (e.g. Lock 1 to Lock 2). This aspect of migration appears comparable to the similar-sized Pacific lamprey in the Basin, USA, which also exhibited lowest rates between release and location of first detection, and similar migration rates to pouched lamprey post first detection (McIlraith et al. 2015). These slower initial rates may reflect recovery from tagging, physiological adaptation following transition from marine to fresh waters (in the case of pouched lamprey in the River Murray), or greater difficulty in navigating upstream through Lake Alexandrina in comparison to linear river reaches.

4.3. Fish passage

Only low numbers of lamprey were detected on fishway PIT readers (Lock 1 = 7, Lock 2 = 2 and Lock 3 = 2), yet all individuals successfully ascended the fishways, with ascent rates predominantly <4 hr. These high passage efficiency and ascent rates are consistent with PIT

24 Bice, C. et al. (2019) Lamprey migration in the River Murray tagged pouched lamprey that ascended fishways on Locks 1–10 in 2015 (Bice et al. 2018), and indicate that the vertical-slot fishways on these weirs are effective in promoting longitudinal connectivity for the species. Nonetheless, our data do indicate that, despite fishways, the weirs on the River Murray may delay the migration of lamprey.

Across 2017 and 2018, single lamprey undertook multiple successful ascents of fishways at Locks 1–3, suggesting fallback over the weir occurred resulting in delays of passing the structure of up to seven days. In addition, in 2018, one individual that eventually passed Lock 1, spent 20 days in the immediate vicinity downstream of Lock 1, prior to locating the fishway. Other individuals that were detected downstream of Lock 1 or Lock 3, but did not enter the fishways, spent 20 hr to 29 days in these areas. Extended periods spent below the weirs, often following rapid migration through downstream reaches, may result from either: 1) difficulty in locating fishway entrances; or 2) searching for or locating suitable holding/spawning locations prior to spawning. Identifying the underlying mechanism is difficult given low numbers of tagged lamprey, but if extended time below weirs is related to difficulty in locating fishway entrances, this migratory delay may represent a substantial impediment to upstream migration. This delay may impact the eventual extent of migration and individual fitness upon reaching spawning grounds. Nonetheless, for PIT tagged only fish in 2015 and 2017, there were many instances of fish being detected on consecutive weir PIT reader systems (up to 88 km apart) within 2 days, suggesting for these fish migratory delay was negligible. Ammocoetes (early life stages of lamprey) have in the past been detected immediately downstream of Lock 1 (SARDI unpublished data), lending support to the hypothesis that considerable time spent downstream of structures may also be related to the location of suitable spawning habitats.

4.4. Future research and management

A key priority for future research is locating regions (i.e. streams and specific river reaches) that provide spawning and nursery habitats for pouched lamprey. The last detection of some individuals entering Currency Creek and the Finniss River suggest that these catchments may provide appropriate spawning sites, whilst the cessation of migration across multiple reaches of the River Murray suggests spawning/nursery sites may be widespread. Nonetheless, confidently identifying such sites relies on detecting ammocoetes, which due to their cryptic benthic habit often requires specialised approaches (Moser et al. 2007). In small clear water streams in New Zealand, that are comparable to the Finniss River, specialised back-pack electrofishing

25 Bice, C. et al. (2019) Lamprey migration in the River Murray techniques are typically used to sample ammocoetes (Jellyman and Glova 2002), but in large turbid rivers like the River Murray, alternative techniques including specialised deep-water electrofishing or dredging may be required (sensu Harris and Jolley 2016). Other ‘passive’ detection techniques, including environmental DNA and ‘pheromone samplers’ have promise in larger systems (Stewart and Baker 2012). We suggest that a combined methodology of active and passive sampling would be appropriate to assess nursery habitat distribution in the southern MDB, particularly targeting tributary streams and tailwaters of weirs in reaches of the lower and mid River Murray.

The ecology and contemporary status of short-headed lamprey in the MDB remains largely unknown. A single short-headed lamprey was sampled, and tagged, in the current project, but was not detected again. This was the first short-headed lamprey sampled entering the MDB since 2011, which in turn was the first verified record of an upstream migrant since 2006 (Bice and Zampatti 2019). Lack of detection in many years may be a function of sampling timing and frequency. Nevertheless, the current status of the species, aspects of migration and interaction with weirs and fishways, and location of spawning/nursery habitats, represent key knowledge gaps for short-headed lamprey.

Data collected in this and allied monitoring projects (e.g. Bice and Zampatti 2019) may be used to inform water and structure management in the southern MDB. Of particular importance is attraction of lamprey to the Murray Mouth and subsequent river ingress. The typical reliance of lamprey on physico-chemical and olfactory cues to stimulate marine–freshwater transition during key periods highlights the importance of maintaining discharge from the Murray Barrages over these periods. In each year of the current study, conspicuous flow events, largely supported by environmental water delivery, occurred in winter (mean daily discharge ranging 1575–16,228 ML.d-1), and pouched lamprey were sampled in association with these events each year. Whilst prescribing precise discharge volumes is difficult with available data, barrage discharge in 2018 was relatively low (mean ± SD = 1575 ± 1899 ML.d-1), but still associated with upstream migration of lamprey. Consequently, in the absence of further data, discharge of this magnitude could be viewed as a tentative minimum barrage discharge target over the migration period.

For both pouched and short-headed lamprey, the use of olfaction to detect cues from conspecifics (e.g. larval pheromones) and guide migration (see Bett and Hinch 2016) warrants further investigation, particularly in regards to the source and longitudinal integrity of riverine flows. Riverine ingress is likely influenced by these pheromones in outflowing water (Meckley et al.

26 Bice, C. et al. (2019) Lamprey migration in the River Murray

2014), but the ultimate extent of riverine migration may also be influenced by conspecific pheromones (Moser et al. 2015). As such, overall river discharge and tributary contributions, as well as densities of ammocoetes in different localities may influence migratory behaviour. In the context of the River Murray, discharge from Mount Lofty tributaries due to rainfall, or from upstream tributaries due to dam releases (e.g. the Goulburn River), and re-regulation of flow through Lake Victoria, may all influence behaviour, migratory extent, and ultimately population dynamics. However, improved knowledge of the influence of flow on migration and population dynamics is first reliant on knowledge of spawning and nursery locations.

Understanding mechanisms of potential migratory delay downstream of lock and weir structures warrants further investigation. The acoustic telemetry technique used provided a means of assessing periods of presence downstream of structures, but did not provide the fine-scale tracking resolution to adequately assess behaviour. Radio telemetry may provide the resolution to investigate behaviour in the immediate vicinity of weirs and fishways (e.g. Keefer et al. 2013), and determine if migratory delay is related to inefficiency in locating fishway entrances or reproduction.

27 Bice, C. et al. (2019) Lamprey migration in the River Murray

5. CONCLUSION This project has improved understanding of the upstream spawning migration of pouched lamprey in the River Murray, particularly extent, timing and rate, as well as interaction with weirs and fishways. These migrations occur over a defined period, with freshwater ingress occurring in July/August, and riverine migration potentially extending to January. The migrations may be long- distance (100s km) and rapid, and are aided by fishways constructed under the Sea to Hume fishway program. Many aspects of pouched lamprey and short-headed ecology remain unknown, but the current project serves to provide direction for future research and management.

28 Bice, C. et al. (2019) Lamprey migration in the River Murray

6. REFERENCES

ABS/ABARE/BRS (2009). Socio-economic context for the Murray-Darling Basin – descriptive report. MDBA publication no. 34/09. Murray-Darling Basin Authority, Canberra.

The Riverine Herald (1950). First Murray lamprey for 20 years. The Riverine Herald. Echuca, Victoria.

Baker, C. F., D. J. Jellyman, K. Reeve, S. Crow, M. Stewart, T. Buchinger and W. Li (2017). First observations of spawning nests in the pouched lamprey (Geotria australis). Canadian Journal of Fisheries and Aquatic Sciences 74: 1603-1611.

Barrett, J. and M. Mallen-Cooper (2006). The Murray River's 'Sea to Hume Dam' fish passage program: Progress to date and lessons learned. Ecological Management and Restoration 7: 173- 183.

Baumgartner, L., B. Zampatti, M. Jones, I. Stuart and M. Mallen-Cooper (2014). Fish passage in the Murray-Darling Basin, Australia: Not just an upstream battle. Ecological Management & Restoration 15: 28-39.

Bett, N. N. and S. G. Hinch (2016). Olfactory navigation during spawning migrations: a review and introduction of the Hierarchical Navigation Hypothesis. Biological Reviews 91: 728-759.

Bice, C. M., M. S. Gibbs, N. N. Kilsby, M. Mallen-Cooper and B. P. Zampatti (2017b). Putting the “river” back into the lower River Murray: quantifying the hydraulic impact of river regulation to guide ecological restoration. Transactions of the Royal Society of South Australia: 108-131.

Bice, C. M., M. P. Hammer, S. D. Wedderburn, Q. Ye and B. P. Zampatti (2018). Fishes of the Lower Lakes and Coorong: an Inventory and Summary of Life History, Population Dynamics and Management. Natural History of the Coorong, Lower Lakes and Murray Mouth region (Yarluwar- Ruwe). L. Mosely, Q. Ye, S. A. Shepard, S. Hemming and R. Fitzpatrick, Royal Society of South Australia, Adelaide: 371-399.

Bice, C. M. and B. P. Zampatti (2019). Fish assemblage structure, movement and recruitment in the Coorong and Lower Lakes in 2017/18, South Australian Research and Development Institute (Aquatic Sciences), Adelaide. SARDI Publication No. F2011/000186-6. SARDI Research Report Series No. 1007. 81 pp.

Bice, C. M., B. P. Zampatti and J. Fredberg (2017a). Assessment of biological effectiveness of newly constructed fishways on the Murray Barrages, South Australia, 2015-2017, South Australian Research and Development Institute (Aquatic Sciences), Adelaide. SARDI Publication No. F2016/000452-2. SARDI Research Report Series No. 952. 65pp.

29 Bice, C. et al. (2019) Lamprey migration in the River Murray

Bice, C. M., B. P. Zampatti and J. F. Fredberg (2016). Fish assemblage structure, movement and recruitment in the Coorong and Lower Lakes in 2015/16 South Australian Research and Development Institute (Aquatic Sciences), Adelaide. SARDI Publication No. F2011/000186-6. SARDI Research Report Series No. 921. 77 pp.

Bice, C. M., B. P. Zampatti and W. M. Koster (2018). Migration ecology of pouched lamprey in the River Murray. Australian Society for Fish Biology Conference. Melbourne, October 2018.

Brownscombe, J. W., E. J. I. Lédée, G. D. Raby, D. P. Struthers, L. F. G. Gutowsky, et al. (2019). Conducting and interpreting fish telemetry studies: considerations for researchers and resource managers. Reviews in Fish Biology and Fisheries 29: 369-400.

Clemens, B. J., T. R. Binder, M. F. Docker, M. L. Moser and S. A. Sower (2010). Similarities, Differences, and Unknowns in Biology and Management of Three Parasitic Lampreys of North America. Fisheries 35: 580-594.

Cooke, S. J., E. G. Martins, D. P. Struthers, L. F. G. Gutowsky, M. Power, S. E. Doka, J. M. Dettmers, D. A. Crook, M. C. Lucas, C. M. Holbrook and C. C. Krueger (2016). A moving target— incorporating knowledge of the spatial ecology of fish into the assessment and management of freshwater fish populations. Environmental Monitoring and Assessment 188: 239.

DeCelles, G. R. and S. X. Cadrin (2010). Movement patterns of winter flounder (Pseudopleuronectes americanus) in the southern Gulf of maine: observations with the use of passive acoustic telemetry. Fishery Bulletin 108: 408-419.

Hammer, M., S. Wedderburn and J. van Weenen (2009). Action Plan for South Australian Freshwater Fishes., Native Fish Australia (SA), 206pp.

Harris, J. E. and J. C. Jolley (2016). Estimation of occupancy, density, and abundance of larval lampreys in tributary river mouths upstream of dams on the , Washington and Oregon. Canadian Journal of Fisheries and Aquatic Sciences 74: 843-852.

Jellyman, D. J. and G. J. Glova (2002). Habitat use by juvenile lampreys (Geotria australis) in a large New Zealand river. New Zealand Journal of Marine and Freshwater Research 36: 503-510.

Jellyman, D. J., G. J. Glova and J. R. E. Sykes (2002). Movements and habitats of adult lamprey (Geotria australis) in two New Zealand waterways. New Zealand Journal of Marine and Freshwater Research 36: 53-65.

Jepsen, N., A. Koed, T. E. B and E. Baras (2002). Surgical implantation of telemetry transmitters in fish: how much have we learned? Hydrobiologia 483: 239-248.

30 Bice, C. et al. (2019) Lamprey migration in the River Murray

Jonsson, B., R. S. Waples and K. D. Friedland (1999). Extinction considerations for diadromous fishes. ICES Journal of Marine Science 56: 405-409.

Keefer, M. L., C. C. Caudill, T. S. Clabough, M. A. Jepson, E. L. Johnson, C. A. Peery, M. D. Higgs and M. L. Moser (2013). Fishway passage bottleneck identification and prioritization: a case study of Pacific lamprey at Bonneville Dam. Canadian Journal of Fisheries and Aquatic Sciences 70: 1551-1565.

Kelso, J. R. M. and G. J. Glova (1993). Distribution, upstream migration and habitat selection of maturing lampreys, Geotria australis, in Pigeon Bay Stream, New Zealand. Australian Journal of Marine and Freshwater Research 44: 749-759.

Maitland, P. S., C. B. Renaud, B. R. Quintella, D. A. Close and M. F. Docker (2015). Conservation of native lampreys. Lampreys: biology, conservation and control. M. F. Docker, Springer. 1: 375- 428.

McDowall, R. M. (1988). Diadromy in Fishes - Migrations between Freshwater and Marine Environments, Croom Helm, London, UK.

McIlraith, B. J., C. C. Caudill, B. P. Kennedy, C. A. Peery and M. L. Keefer (2015). Seasonal Migration Behaviors and Distribution of Adult Pacific Lampreys in Unimpounded Reaches of the Snake River Basin. North American Journal of Fisheries Management 35: 123-134.

MDBA (2009). The Basin Plan: a concept statement. Murray-Darling Basin Authority, Canberra.

MDBA (2014). Basin-wide environmental watering strategy, Murray-Darling Basin Authority, Canberra. MDBA Publication No. 20/14.

Meckley, T. D., C. M. Wagner and E. Gurarie (2014). Coastal movements of migrating (Petromyzon marinus) in response to partial pheromone added to river water: implications for management of invasive populations. Canadian Journal of Fisheries and Aquatic Sciences 71: 533-544.

Moser, M. L., P. R. Almeida, P. S. Kemp and S. P. W (2015). Lamprey spawning migration. Lampreys: biology, conservation and control. M. F. Docker, Springer. 1: 215-263.

Moser, M. L., J. M. Butzerin and D. B. Dey (2007). Capture and collection of lampreys: the state of the science. Reviews in Fish Biology and Fisheries 17: 45-56.

Potter, I. C. (1970). Life cycles and ecology of Australian lampreys of the genus Mordacia. Journal of Zoology 161: 487-511.

31 Bice, C. et al. (2019) Lamprey migration in the River Murray

Potter, I. C. and F. L. S. Strahan (1968). The of the lampreys Geotria and Mordacia and their distribution in Australia. Proceedings of the Linnean Society of London 179: 229-240.

Quinn, T. P. and D. J. Adams (1996). Envoironmental changes affecting the migratory timing of American shad and sockeye . Ecology 77: 1151+.

Stewart, M. and C. F. Baker (2012). A Sensitive Analytical Method for Quantifying Petromyzonol Sulfate in Water as a Potential Tool for Population Monitoring of the Southern Pouched Lamprey, Geotria Australis, in New Zealand Streams. Journal of Chemical Ecology 38: 135-144.

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

32 Bice, C. et al. (2019) Lamprey migration in the River Murray

7. APPENDIX Appendix 1. Biological and movement details of pouched lamprey captured and tagged in 2016, 2017 and 2018. Tagging location: BCB = Boundary Creek Barrage, GB = Goolwa Barrage, MB = Mundoo Barrage, TB = Tauwitchere Barrage. Tag type: P = PIT tag, A = acoustic tag). *signifies the last detection was on a fishway PIT reader system. ND = not detected.

Fish Tagging Tagging Tag type Acoustic Total length Weight Time at Last detection Distance no. date location ID (mm) (g) liberty (d) location from release (km) 2016 1 18/08/2016 GB P - 521 - - ND - 2 18/08/2016 BCB P - 527 - - ND - 3 30/08/2016 GB P - 562 - - ND - 4 30/08/2016 GB P - 545 - - ND - 5 31/08/2016 GB P - 588 - - ND - 6 31/08/2016 GB P - 580 - - ND - 2017 1 05/07/2017 TB P + A (V7) 1331 580 194 15 Clayton 28.5 2 06/07/2017 GB P + A (V7) 1332 521 173 4 Clayton 23.5 3 06/07/2017 GB P + A (V7) 1333 573 159 1 Clayton 23.5 4 07/07/2017 MB P + A (V7) 1334 603 208 15 Clayton 17.5 5 07/07/2018 MB P + A (V7) 1335 630 247 23 Wellington 17.5 6 18/07/2017 TB P + A (V7) 1336 593 177 - ND - 7 19/07/2017 MB P + A (V7) 1337 527 136 13 Clayton 17.5 8 20/07/2017 MB P + A (V7) 1338 541 160 1 Clayton 17.5 9 21/07/2017 TB P + A (V7) 1339 557 170 14 Mannum 150

33 Bice, C. et al. (2019) Lamprey migration in the River Murray

Fish Tagging Tagging Tag type Acoustic Total length Weight Time at Last detection Distance no. date location ID (mm) (g) liberty (d) location from release (km) 2017 10 21/07/2017 TB P + A (V9) 53027 599 187 13 Mannum 150 11 21/07/2017 TB P + A (V9) 53028 577 190 - ND - 12 21/07/2017 TB P + A (V9) 53029 625 191 - ND - 13 1/08/2017 TB P + A (V9) 53030 598 203 4 Clayton 28.5 14 1/08/2017 TB P + A (V7) 1340 568 173 - ND - 15 2/08/2017 MB P + A (V7) 1347 555 155 27 Goolwa Channel 30.5 16 2/08/2017 TB P + A (V7) 1341 524 146 - ND - 17 2/08/2017 MB P + A (V7) 1342 547 168 - ND - 18 2/08/2017 MB P + A (V9) 53031 570 181 - ND - 19 2/08/2017 MB P + A (V7) 1343 533 146 - ND - 20 2/08/2017 MB P + A (V9) 53033 570 170 - ND - 21 2/08/2017 MB P + A (V7) 1344 505 132 - ND - 22 2/08/2017 MB P + A (V7) 1345 555 167 - ND - 23 2/08/2017 MB P + A (V7) 1346 516 148 - ND - 24 2/08/2017 MB P + A (V9) 53034 580 186 - ND - 25 2/08/2017 MB P + A (V9) 53032 568 173 16 Wellington 74 26 3/08/2017 TB P + A (V7) 1348 537 148 1 Finniss River 33.5 27 3/08/2017 TB P + A (V9) 53035 560 156 - ND - 28 3/08/2017 MB P + A (V9) 53036 585 183 - ND -

34 Bice, C. et al. (2019) Lamprey migration in the River Murray

Fish Tagging Tagging Tag type Acoustic Total length Weight Time at Last detection Distance no. date location ID (mm) (g) liberty (d) location from release (km) 2017 29 3/08/2017 MB P + A (V9) 53037 556 181 - ND - 30 3/08/2017 MB P + A (V9) 53038 558 151 - ND -

31 3/08/2017 MB P + A (V9) 53039 548 157 - ND -

32 4/08/2017 TB P + A (V9) 53040 590 185 28 Lock 1* 274

33 4/08/2017 TB P + A (V9) 53041 610 175 12 Wellington 74

34 15/08/2017 GB P + A (V7) 1335 523 132 4 Clayton 23.5

35 15/08/2017 GB P + A (V7) 1350 487 105 11 Finniss river 30

36 15/08/2017 GB P + A (V7) 1349 535 124 0 Goolwa Channel 10.5

37 15/08/2017 MB P + A (V9) 53028 545 168 - ND -

38 15/08/2017 TB P + A (V9) 53042 541 141 - ND -

39 29/08/2017 GB P + A (V9) 53043 588 201 - ND -

40 31/08/2017 GB P + A (V7) 38701 518 138 5 Currency Creek 19.5

41 31/08/2017 GB P + A (V9) 53044 610 179 2 Finniss River 30

42 12/09/2017 GB P + A (V7) 38702 535 134 19 Finniss River 30

43 14/09/2017 GB P + A (V7) 38703 518 124 16 Clayton 23.5

44 06/07/2017 MB P - 510 - - ND -

35 Bice, C. et al. (2019) Lamprey migration in the River Murray

Fish Tagging Tagging Tag type Acoustic Total length Weight Time at Last detection Distance no. date location ID (mm) (g) liberty (d) location from release (km) 2017 45 20/07/2017 MB P - 525 134 - ND -

46 20/07/2017 TB P - 470 131 - ND -

47 21/07/2017 TB P - 572 196 30 Lock 1* 274

48 02/08/2017 MB P - 538 152 33 Lock 2* 362

49 02/08/2017 MB P - 577 190 76 Lock 3* 431

50 03/08/2017 TB P - 541 152 - ND -

51 15/08/2017 GB P - 514 122 - ND -

52 17/08/2017 TB P - 550 147 26 Lock 1* 274

53 17/08/2017 TB P - 520 113 - ND -

2018 1 11/07/2018 TB P + A (V7) 9652 579 190 19 DS Lock 1 274 2 13/07/2018 MB P + A (V7) 9653 581 185 95 DS Lock 3 431

3 26/07/2018 GB P + A (V7) 9654 515 146 - ND -

4 27/07/2018 GB P + A (V7) 9655 600 200 54 Lock 3* 431

5 27/07/2018 GB P + A (V7) 9656 573 156 25 DS Lock 1 274

6 27/07/2018 GB P + A (V7) 9657 535 137 14 Clayton 23.5

36