Jim Barrett is Convenor of the Fish Passage Task Force and Manager of the Native Fish Strategy, Murray-Darling Basin Commission (GPO Box 409, Canberra, ACT 2601, Australia, Email: email@example.com). Martin Mallen-Cooper is the Principal Consultant of Fishway Consulting Services (8 Tudor Pl., St Ives, NSW 2075, Australia, Email: firstname.lastname@example.org) and a member of the Fish Passage Task Force.
Since its commencement in 2001, a program to facilitate fish passage on a major stretch of Australia's longest river has installed eight structures, testing and modifying their design as they go. What are the results so far and what are the implications for future directions?
Across the Murray-Darling Basin thousands of weirs obstruct the passage of fish and contribute to the significant declines in distribution and abundance suffered by many fish species. As part of an ambitious plan to rehabilitate native fish populations, the Murray-Darling Basin Commission (MDBC) is restoring fish passage along the Murray River from the sea upstream to Hume Dam – a distance of 2225 km. The program, instigated in 2001, is a multistate process involving engineers and fish biologists in the design, construction, testing and evaluation of fishways at 15 weirs and barrages along the main stem of the river. The program incorporates adaptive management, investigating innovation and rapidly applying research results. This article provides information to date on the successes of the program and lessons learned during the first five years.
Why the need for fishways?
All fish need to move along streams for feeding, spawning, to seek shelter and refuge, for dispersal of young fish, to counter downstream displacement in high flows, and to recolonize after droughts. Almost every aspect of the life history of fish involves movement and this can be over scales of metres or hundreds of kilometres (Lucas & Baras 2001). These movements can be regular seasonal migrations undertaken by much of the population or they can be less regular and less well defined such as dispersal or recolonization movements which may be undertaken by large components of the fish community or by a few individuals (Northcote 1998; Mallen-Cooper 2000). Movement is part of major ecological processes and preventing or inhibiting movement has a range of impacts, from increased predation by birds below barriers to local extinctions of native species. In the Native Fish Strategy for the Murray-Darling Basin 2003–2013 (MDBC 2004), barriers to fish passage are identified as one of eight key threats to native fish populations. This is not unusual in regulated rivers throughout the world where lack of fish passage is frequently identified as a major cause of the declines of freshwater fish (Northcote 1998).
The fish community of the lower Murray River is listed as a threatened ecological community in New South Wales. It contains a range of larger fish species such as Murray Cod (Maccullochella peelii peelii), Golden Perch (Macquaria ambigua) (Fig. 1), and the threatened Silver Perch (Bidyanus bidyanus) and 11 small fish species (ranging from 20 to 70 mm in size) with five of these known to specifically move upstream (Table 1, Fig. 2). Of the small fish, three are classified as threatened: Murray Hardyhead (Craterocephalus fluviatilis), Olive Perchlet (Ambassis agassizii) and Flat-headed Galaxias (Galaxias rostratus).
The objective of providing fish passage in the Murray River is to directly ameliorate these impacts and make a major contribution to the rehabilitation of native fish by restoring stream continuity and migration pathways.
Early Australian fishways and their design evolution
Fishways have been present on the Murray River since the 1930s, using designs imported from northern hemisphere prototypes (Box 1). Two early fishways were incorporated into the design and construction of Weir and Lock No 6 (Murtho) in 1930 and No 15 (Euston) in 1937 (Fig. 3). However, these ‘submerged–orifice’ designs, with respective gradients of around 1-on-6 and 1-on-9 (i.e. 1 m of vertical rise for 9 m of horizontal length), were designed to the known standards of the day which applied to the powerful upstream swimming abilities of Northern Hemisphere salmon. Performing in an Australian context, however, these fishways provided very limited passage for native fish.
Later observations, including laboratory experiments in the 1980s (Mallen-Cooper 1994), gave a sounder understanding of swimming abilities and behaviour of native fish in fishways. With the Commission agreeing to fund a fishway at Torrumbarry Weir, this new biological information was applied to its design, and construction was undertaken in 1990. This was a ‘vertical–slot’ design (Fig. 5) which had a gradient of 1-on-18.
The operation of this structure proved to be a major advancement for fish passage in the Murray River. Accumulations of native fish downstream of the weir (an indicator of obstructed fish passage) declined dramatically following its introduction, with over 20 000 fish recorded as moving through the fishway in 3 years, including smaller fish (Mallen-Cooper 1999). At the same time, native fish, particularly Silver Perch and Golden Perch, reappeared in significant numbers in the reaches of the Murray River upstream of Torrumbarry. Following Torrumbarry, other vertical-slot fishways were built elsewhere in Australia which increased the knowledge of the diversity and size range of fish that could use this type of fishway design (e.g. Stuart & Mallen-Cooper 1999).
The total reconstruction of Torrumbarry Weir in 1994–1996 included the decommissioning and replacement of the fishway. Further knowledge gained during the period that the first fishway operated was applied to the new design.
The Sea to Hume Dam program
How it started
The vision of establishing fish passage along the Murray River was first articulated in the mid-1980s when the MDBC (then River Murray Commission) established a working group of interstate scientists to examine the issue of fish passage. The working group prioritized all the weirs on the main stem of the Murray River, based on their ecological impact. As the first priority fishway, Torrumbarry fishway was constructed in the early 1990s. The next fishway was installed at Yarrawonga Weir, the next barrier upstream, which coincided with the construction of a power station at the site. Contemporary world practice for an 8-m-high weir suggested that this should be a fish lock. Although this was the first fish lock in the Murray-Darling Basin, the experience gained at Torrumbarry on fish behaviour and swimming abilities provided direct design input. These two fishways opened up 1115 km of river from Lock 15 (Euston) to Hume Dam.
The genesis of the present program (involving a wholly funded fish passage strategy from Hume Dam to the sea) was initially triggered by works required to comply with new Occupational Health and Safety (OH&S) standards at Murray River weirs. In 2000, River Murray Water (RMW) made a decision to refurbish Locks 1–11 and 15, covering a continuous river length of 1110 km. This drew the issue of fish passage on these structures to the attention of Ministers under relevant State legislation, with the result that, in March of 2001, the Murray-Darling Basin Ministerial Council adopted a Murray River structural works program to provide passage for migratory fish from the sea to Hume Dam, a distance of 2225 km. The program included the tidal barrages at the Murray mouth, which then provided the continuous length of river connected with the sea. This 8-year program was estimated to cost a total of $45 million.
At its inception the MDBC formed the Fish Passage Task Force (FPTF) to provide specialist technical advice on the fishways and to provide continuity for a lengthy program. The FPTF comprises fish scientists and engineers from across the jurisdictions in the basin, along with the owner and operator of the structures. The group's primary role is to guide fishway design and evaluate the success of fish passage in the Sea to Hume Dam program, with a long-term, broader role of developing a basin-wide fish passage program.
Setting design criteria for the fishways
The first challenge of the FPTF was to set ecological objectives for fish passage for a whole-of-ecosystem approach to a river reach with high conservation value; from these followed design criteria for the fishways. Biology and engineering are inherently linked in developing fishway design. The biological challenge is to build a model of fish migration, identifying species and life stage, seasonality of movement, response to flow, and interaction with river hydrology. From that model, engineering and biology need to interact as the cost-sensitive areas of the design are identified and design decisions can be made on an ecological basis.
Studies to date have shown that the migratory response of native fish species is related to season and river flow (O’Connor et al. 2005), which creates migratory pulses (McDowall 1996; Mallen-Cooper 1999). Most fish move in spring and summer, tapering off in early autumn. Within these seasons, Golden Perch and Silver Perch are stimulated to move by increases in river flow. Immature and mature life stages of these species also respond differently to varying increases in river flow. Small fish species, such as Australian smelt, are moving mainly at the stable lower flows while larger species such as Bony Herring and the non-native Common Carp (Cyprinus carpio) are moving upstream in relation to increased temperatures which can occur over a wide range of flows (Mallen-Cooper 1999).
Significant ecological information gaps still remain, however, requiring precautionary design decisions to be included. For example, little is known about the contribution that migratory adult Murray Cod make to recolonization and recruitment of the population as a whole, compared with adult fish that may not be migratory. The FPTF concluded that the fishways should enable movement of a wide size range of adult Murray Cod. The challenge for fishway design is therefore to incorporate present data, identify knowledge gaps and include some flexibility in the fishway design.
For fishway design, the migratory fish can be grouped into three major size classes: 20–70 mm (e.g. Australian Smelt, Flyspecked Hardyhead, Gudgeon species); 90–600 mm (immature and mature Silver Perch and Golden Perch, Bony Herring, immature Murray Cod); and 600–1400 mm (adult Murray Cod). Providing passage for the full size range of fish would have increased cost to a prohibitive extent. Designing for the smallest fish would need very low water velocities and a long fishway while designing for the largest fish would necessitate very large fishway pools; such parameters have a major influence on cost. The FPTF therefore identified a goal of designing fishways to allow passage at weirs for a substantial proportion of all species and life stages of fish.
A minimum size criterion of 40–60 mm was established for the fishway designs, which is intended to enable upstream passage of approximately half of the fish between 20 and 70 mm, with the ecological objective of enabling dispersal of these species. Passing all fish between 90 and 600 mm (which includes species with well-known migratory patterns) was identified as a goal. A maximum size criterion of 1000 mm was intended to enable the upstream passage of the majority of the adult Murray Cod.
The range of river flows over which a fishway operates has a major bearing on its cost. The FPTF considered that the fishways needed to operate over a wide range of flows. As different species and life stages of fish migrate at different flows, reducing the operational range could have greatly impacted some life stages of some species.
These biological and hydrological criteria determine specific aspects of the fishway design, and are adjusted according to the dimensions and configurations of the individual locks and weirs, which are also specifically managed to attract fish to the fishway by adjusting flow patterns (Fig. 4). For the Murray River, the vertical-slot design (Fig. 5) was used on a 1-on-32 gradient with 3-m long by 2-m wide pools. The step height (head loss) between pools is 0.1 m, which creates a maximum water velocity of 1.4 m s−1; combined with the pool volume this generates a low turbulence of 40 W/m3. The pool size enables the passage of Murray cod and the low water velocities and turbulence are intended for the small fish. The fishways also have rocks glued to the floor to enhance the passage of crustaceans (Fig. 6).
Carp management in fishways
Concentrations of carp at fishways provide opportunities for their removal and facilities need to be included in the design phase. To date removal has been through labour intensive trapping that places considerable stress on native fish. The ‘Williams’ carp separation cage’ makes use of the differing behaviours of carp and native fish (Stuart et al. in press) resulting in effective separation of carp. The cage is low-cost, is being designed to be included in the new and existing fishways and is likely to have application throughout fishways in the southern Murray-Darling Basin.
Where are they being built?
The Sea to Hume Dam program has been underway since July 2001. It aims to deliver 15 effective fishways at 13 sites from the tidal barrages near the Murray Mouth to Weir and Lock 15 at Euston. Eight of the planned fishways have been installed; five at the riverine locks (Table 2) and three at the Barrages (Table 3).
Table 2. Progress towards construction of fishways at Locks and Weirs
Year of completion
Lock & Weir No.
Completed (Denil fishway)
Designed, ready for tender
3, 5, 11
Table 3. Progress towards construction of fishways at the barrages
Year of completion
Type of fishway
Vertical-slot and rock ramp fishways
Three remaining barrages
Up to 13 fishways may be needed, dependent on results from Goolwa and Tauwitchere
At each site, the FPTF considers several options for the location, design and number of fishways. At the riverine locks, fish passage along both banks was preferred, but as this increased costs significantly it was considered that, by manipulating flows across the weir, a single fish passage was adequate. At the tidal barrages, assessment showed that all five weirs impede the passage of fish, but that low-cost, effective fishways are feasible because the difference in water levels either side is very small. The two barrage sites where fishways have been completed were identified as the highest priorities for intervention because of the high biomass of fish present.
A schedule is proposed that will see the construction of fishways at all eight remaining structures by 2010. The program will eventually be extended to other major obstructions to fish movement throughout the Murray River system, and includes the decommissioning of redundant structures.
The vertical-slot fishways presently being installed at the Locks on the Murray River are world-class in functionality. However, their broad-scale application throughout the Basin is likely to be costly and therefore unlikely to be practical. Even with the additional cost of retrofitting the fishway to an existing River Murray Lock and Weir, the cost (about AUD$3m) represents only about 7% of total estimated replacement cost of the structure. In 2004, the FPTF conducted a workshop, facilitated by the Australian Centre for Value Management, to explore opportunities to reduce costs of fishway design and construction while ensuring that functionality and other benefits were preserved. Several cost-saving fishway configurations were identified, and a recommendation made that a research and development program be developed to explore additional options.
A major factor in the cost is the slope of the fishway channel, which determines the total length of the fishway. Hence, an experiment was conducted at the Lock 8 fishway in 2005 to compare the performance of variations in the vertical-slot baffle at higher slopes. One finding was that separating the function of passing small fish and large fish is likely to reduce cost and improve performance; that is, two fishways could be cheaper than one. This significant new knowledge is being applied to the design of an inexpensive fishway for small native fish. Small Denil fishways have been used for small fish in Japan and these are now being tested for small native fish in the Murray River. An experimental Denil fishway for larger native fish was installed at Lock 15 in the existing 1-on-9 fishway channel to improve fish passage until the program reaches the site in 2008 (Fig. 7).
Performance of the new fishways
The tri-State research team (see monitoring, Box 2) has completed its assessment of the functionality of the Lock 8 fishway, the first of the main-stem fishways, and collected 12 fish species and 27 215 individuals in 40 days of sampling. Fish successfully reaching the exit ranged in size from 40 mm long gudgeons to 1030 mm Murray Cod and predation from birds has dramatically reduced. The functionality of the fishway meets the initial design specifications but the biomass of very small fish (20–40 mm long – a size that was outside the original target range) was greater than expected. The tri-State team has now successfully completed a range of innovative fishway experiments aimed at facilitating passage of these fish and improving passage of invertebrates. One of these experiments is using the existing vertical-slot fishway as a fish lock, which has shown that the passage of small fish such as Flyspecked Hardyhead, Gudgeons, Australian Smelt and Murray Rainbowfish, as well as freshwater prawns, can be greatly improved.
Beyond the program
The majority of actions addressing fish passage problems have focused on provision of upstream migration of fish, but there are major issues associated with the impediments to downstream migration of fish. An FPTF workshop in 2003 (Lintermans & Philips 2003) engaged stakeholders in a review, update and prioritization of the actions necessary to restore downstream fish migration, and the need to develop an action plan that will guide investment in this area over the next three to five years. Major issues associated with downstream migration include:
• Inability of adults to return to home-sites after upstream spawning migrations or other movement
• Depletion of downstream fish populations as barriers restrict colonization from upstream areas
• Injuries to fish after passing through or over weirs and spillways
• Larval fish settling out in unsuitable weir-pool habitats
• Fish being diverted into irrigation channels and inappropriate habitats
• Fish being diverted into off-stream storages, turbines, pumping inlets, etc.
Identification of these issues has led the MDBC to fund a project investigating downstream fish movements into irrigation offtakes and pumps.
Extending the program beyond the Murray River
It is anticipated that the program of fish passage will be extended to other major barriers throughout the Murray River system; that is, structures associated with auxiliary river channels (e.g. Mullaroo Creek and the Chowilla system, Lake Victoria and the Menindee Lakes). It will also consider works to Stevens Weir on the Edward River. The MDBC is also addressing the issue of redundant structures, including a review of criteria for assessing structures for decommissioning. Other than the Murray River system, there are nearly 4000 registered weirs throughout the rest of the Murray-Darling Basin. The vast majority of these are low-level weirs (<3 m), requiring relatively simple fishways or fish passage solutions. The FPTF has developed criteria to prioritize fishway placement at these barriers across the basin, based on rankings of the characteristics of the fish species present, including their threatened status, the quality of fish habitat present and economic considerations. The Task Force identified the 20 priority sites in New South Wales, South Australia and Victoria most in need of new fishways or modifications to existing structures (Table 4).
Table 4. Barriers needing new fishways or modifications to existing fishways, listed in order of priority
Menindee Lakes Main Weir
Mullaroo control structure
Lindsay River–Mullaroo Ck
Tea Garden Creek Weir
Neil Turner Weir
A strength of the ‘Sea to Hume Dam’ fish passage program has been the establishment of the FPTF and the tri-State assessment team. Biologists, engineers and river operators working together have ensured the fishways have been designed with transparent ecological goals; and the existence of the FPTF enables the continuity of experience and expertise throughout this long-term program. The interaction of scientists from three states in the assessment program has been a key to achieving a high level of research and the performance of the fishways has provided further confidence for state agencies to install fishways.
The construction of fishways is but one element of the Native Fish Strategy of the MDBC (MDBC 2004) – with other elements including protecting and rehabilitating fish habitat and threatened species, controlling alien fish species, protecting threatened native fish species, managing fish translocation and stocking and communicating the program to the public (Box 3). Providing fish passage on the Murray River is a very tangible and significant step towards the strategy's goal of restoring native fish populations.
Box 1. International contextFishways are widespread throughout the developed world where there are barriers in rivers and where fish resources are valued (Clay 1995). There are extensive fish passage projects in North America, notably the well-known Columbia/Snake river system (Williams 1998) in the north-west and the Connecticut (Moffit et al. 1982) and other rivers in the north-east. Europe has a long history of fishways and there has been a resurgence in this area in the last two decades (Cowx & Welcomme 1998; FAO/DVWK 2002; Larinier et al. 2002) with examples on large rivers like the Rhine (Gebler 1998), Danube (Mader et al. 1998) and Garonne/Dordogne rivers (Travade et al. 1998). Fishways are also widespread in South America (Quirós 1989) and the former USSR (Pavlov 1989). There are, however, some major differences between these projects and the fishways on the Murray River.The ‘Sea and Hume Dam’ program is one of the longest fish passage restoration projects, in terms of river length, in the world. This is achievable in part because of the low gradient of the river, which means that all the weirs from the Barrages to Lock 15 inclusive are less than 5-m high, with most less than 3.5 m. The overseas examples mentioned above generally have steeper gradients, some having much higher fishways passing dams of 5–20-m high; making the achievement of fish passage more difficult.The other major difference between Australian fishways and those in many other countries is the comprehensive ecological objectives of the Murray River program. With the exception of Germany and Austria where recent fish passage projects often have holistic goals (Waidbacher & Haidvogl 1998; FAO/DVWK 2002), fishways in the northern hemisphere (Clay 1995) and South America (Quiros 1989) tend to be targeted to large-bodied species, with the movement of other species considered incidental. In contrast, the Murray River program aims to optimize the passage of the whole fish community and has a very wide target size range (40–1000 mm) compared with fish passage projects in other countries.
Box 2. MonitoringThe FPTF identified desirable performance outcomes for the fishways program. In brief these were:• To pass a large biomass of fish comprised of all native fish species and all fish sizes, by voluntary fish movement primarily aimed at upstream fish passage, but with downstream passage accommodated where feasible• To provide fish passage for close to 100% of the time during the fish migrating months of August to April and to continue operating during environmental weir pool raisings and lowerings• To provide fishways of long life (design life = 100 years), and• To accommodate appropriate monitoring and carp separation where practicalTo monitor and assess the outcomes of the project the MDBC, in 2001, established a team of freshwater scientists from the states of New South Wales, Victoria and South Australia. This tri-state research team is conducting a comprehensive research program that is sampling fish populations in the river (before and after installation of the new fishways) and monitoring fish as they approach, pass through, and leave the fishways (Stuart et al. 2004). Four techniques are providing data on the effectiveness of the newly installed fishways: (i) electrofishing below Locks 1–3; (ii) Passive Integrated Transponder (PIT) tags at fishways; (iii) direct sampling of the fishways; and (iv) pre- and postconstruction comparison at the barrages. The results are informing the design and placement of future fishways, and guiding improvements to newly installed fishways.PIT tags are small glass encapsulated tags with an individual code that are inserted into fish and can be decoded when the fish passes a reader. This type of electronic monitoring enables the identification of (i) large-scale migrations of fish; and (ii) passage time through fishways once the fish have successfully located the entrance.Since its inception, the tri-state team has implanted 15 000 fish, including Common Carp, Golden Perch, Bony Herring, Goldfish, Murray Cod, Silver Perch and Redfin Perch. An additional 5000 tagged fish are also at liberty from other research programs within the Murray-Darling Basin. PIT readers are now operating at Locks 7, 8, 9, 10, 15 (Euston), 26 (Torrumbarry) and Yarrawonga. At all newly completed fishways antennas have been installed at the entrance, exit and two midpoints to provide detailed information regarding the behaviour of large-bodied fish within the fishways. This information is continuously logged and the data is automatically relayed to the researchers weekly.In 2003–2004 the MDBC funded the CSIRO to test a split-beam hydroacoustic system for counting fish. The results demonstrated that it is possible to estimate total abundance and size structure of fish in large rivers and at a large scale; count fish passing through fishways in both directions; estimate fish size and determine fish density, speed of movement and coordinates in space even in moderately turbulent creeks as shallow as 1 m.Trials are underway to test new technologies for remote sampling of fish including an underwater Dual-frequency IDentification SONar (DIDSON) system for remotely studying the migration patterns of native fish. The DIDSON system uses acoustic lenses to produce high-resolution video and is particularly useful in dark and turbid conditions where conventional optical systems are ineffective. Trials of the DIDSON at Yarrawonga Weir, Lock 8 and the Murray mouth barrages revealed detailed aspects of fish behaviour at close range that have immediate implications for fishway design and for interpreting the results of monitoring. Further trials of the unit are currently being planned.
Box 3. Communication to the publicTo raise community awareness of fish passage issues in the basin, and to increase understanding of the purpose and science behind fishways, the FPTF has made extensive use of the media and has been involved in the development of educational material. Examples include the brochures: ‘What is a fishway?’, an explanation of the principles behind fishways and justification for their use; the recent ‘The River Murray Barrages and Fishways’, which describes the ecology of fish species in the Lower Lakes and Coorong, as well as the construction of fishways at the Tauwitchere and Goolwa barrages; and ‘Fish n’ Chips’, a description of the application of PIT tagging technology to assess fish movements and fishway effectiveness in the basin. The FPTF is also developing a brochure to inform the public on activities associated with the upgrading of locks and weirs.Interpretive signs have also been developed for fishways, both completed and under construction, at Goolwa and Lock 9. These signs contain information about the design and purpose of the fishway, an explanation of the ‘Sea to Hume Dam’ fish passage program, and list the species expected to pass through the fishway.A video/DVD has also been produced which gives an overview of the basin and its native fish, describes the need for fish passage and fishways, and gives the viewer a ‘fish's eye’ view of an animated vertical-slot fishway. This has proved extremely popular at the numerous events that have been attended by the Native Fish Strategy team. A similar project is underway to animate the rock-ramp fishway at Tauwitchere barrage, to complement the audio-visual material available.The MDBC is committed to the continued production of high quality publications that inform and educate all stakeholders and the wider community on fish passage in the basin. Planned activities include the production of a regular newsletter updating progress with the Sea to Hume Dam fish passage program and a brochure on methods of tracking fish movements, as well as further development of the NFS website to include regularly updated results of the fishways monitoring program. Publications can be obtained from http://www.mdbc.gov.au.
Thanks is extended to the members of the Fish Passage Task Force: John Prentice (RMW), Cameron Lay, Lee Baumgartner (NSW DPI (Fisheries)), Tim O’Brien and Ivor Stuart (VIC DSE (ARI)), Brenton Zampatti (SARDI), Andrew Berghuis (Queensland DPI & Fisheries), Brian Cooper (NSW Department of Commerce), Brenton Erdmann (SA DWLBC), Darryl Jones (SA Water) and Angelo Rossi and Robert Alaia (SA Water Project Managers). David Dreverman (General Manager River Murray Water) made useful comments on the draft text.