Rehabilitation of Waurn Ponds Creek, Victoria, for Yarra Pygmy Perch (Nannoperca obscura) and Other Native Freshwater Fishes A report to Corangamite Catchment Management Authority Paul Close, Erin Webb & Wayne Koster Freshwater Ecology Arthur Rylah Institute for Environmental Research July 2002 ©State Government, Victoria, Department of Natural Resources and Environment

TABLE OF CONTENTS

1. INTRODUCTION

2. METHODS

2.1 STUDY AREA

2.2 STUDY SITES

2.3 OUANTIFICATION OF HABITAT ATTRIBUTES

2.4 SURVEY METHODOLOGY

3. RESULTS

3.1 WATER QUALITY PARAMETERS

3.2 SITE CHARACTERISTICS

3.3 FISH SURVEYS

4. DISCUSSION

4.1 SPECIES DIVERSITY, DISTRIBUTION AND ABUNDANCE

4.2 SIGNIFICANT SPECIES

4.3 THREATENING PROCESSES IMPACTING ON AQUATIC FAUNA IN WAURN PONDS CREEK.

5. ACKNOWLEDGEMENTS

6. REFERENCES

1. INTRODUCTION

As part of the Waurn Ponds Creek rehabilitation project, the Corangamite Catchment Management Authority (CCMA) commissioned the Freshwater Ecology Section of the Department of Natural Resources and Environment to undertake an assessment of the current ecological status of fish assemblages in Waurn Ponds Creek. Waurn Ponds Creek is a short tributary stream of the lower reaches of the Barwon River in southwest Victoria. The catchment is highly modified from its natural state by agricultural development in the mid to upper reaches and urban development in the mid to lower reaches. Increases in urbanisation and the intensification or diversification of agriculture may all represent substantial risk to the integrity and sustainability of biodiversity in aquatic ecosystems (Klein 1979; Growns et al 1998; Sonneman et al 2001; Wash et al 2001; Close 2001).

Despite this, the intrinsic value of streams that flow through agricultural and urban areas is high and there is substantial effort increasingly being afforded to rehabilitate these systems. This is especially true for Waurn Ponds Creek which has potentially high aesthetic and biodiversity values in the mid to lower catchment and high resource values in the upper catchment with regard to supply of water for agricultural demand (pers. observ). Nevertheless, limited information on the ecology of urban and metropolitan watersheds has until recently hindered rehabilitation efforts.

Biodiversity values in Waurn Ponds Creek may include potentially high diversity of fish species including populations of several threatened freshwater fish, namely Yarra pygmy perch (Nannoperca obscura) and Australian mudfish (Neochanna cleaveri). Due to the small size of these and other native fish species that typically inhabit coastal drainages in southeast Australia such as Waurn Ponds Creek, their requirements in rehabilitation plans are not often considered. However, small freshwater fish species often comprise the majority of species diversity in such systems. An understanding of their distribution, abundance and viability throughout the region is therefore crucial to sustaining native fish species diversity through stream rehabilitation efforts.

This study aimed to provide information on the diversity, distribution and abundance of freshwater fish species in Waurn Ponds Creek. In particular, the distribution and population dynamics of Yarra pygmy perch and Australian mudfish were investigated. We have used knowledge of fish assemblages combined with the identification of anthropogenic factors potentially impacting on freshwater fish to set priorities for the design and establishment of a rehabilitation plan for Waurn Ponds Creek.

2. METHODS

2.1 STUDY AREA

Waurn Ponds Creek is a short low altitude tributary of the lower reaches of the Barwon River. From its source at 160 metres above sea level the creek flows generally east for approximately 20 kilometres towards its confluence with the Barwon River. In the upper reaches the creek flows through undulating coastal hill slopes which have been mostly cleared for agricultural purposes. In this region the creek is characterised by cleared riparian vegetation and degraded instream habitats. In the lower reaches the catchment is urbanised and the stream channel has been artificially incised and straightened to mitigate against flooding. In this region riparian vegetation has been mostly cleared although substantial replanting of native riparian vegetation has occurred between Grovedale and Barwon Heads road. Numerous factors impacting on environmental condition and more specifically on aquatic fauna assemblages in the catchment were identified in this study. In general these disturbances relate to the entire catchment and include:

• altered flow regimes;
• degradation of riparian (streamside) vegetation;
• degradation of instream habitat;
• channel modification and;
• instream barriers.

2.2 STUDY SITES

A total of nine survey sites were selected within the study region (Table 2-1). All sites were selected to represent the available habitat types within the catchment. Five sites were located in urban reaches and four sites were located in predominantly agricultural reaches. At each site, a single survey reach was established to represent the available habitat attributes including flow types and cover elements. The location of each survey reach was defined by Australian Map Grid Reference (AMG) from topographic 1:1 00,000 scale maps (Table 2-1).

2.3 QUANTIFICATION OF HABITAT ATTRIBUTES

2.3.1 Hydrology

2.3.2 Cover Elements

2.4 SURVEY METHODOLOGY

2.4.1 Fish Collection

2.4.2 Sample Processing

3. RESULTS

3.1 WATER QUALITY PARAMETERS

3.2 SITE CHARACTERISTICS

Site 1- Henty Main Road

Site 2- downstream of Anderson Road Crossing

This reach extended 100m downstream from Anderson Road Crossing. The mean width and depth were 2 m and 0.1 m respectively. Hydraulic habitats included pool (80%) and run (20%). The streambed comprised predominantly fine substrata including sand (20%), silt (40%) and clay (40%). Instream habitat available as cover to aquatic fauna was dominated by aquatic vegetation (80%). Streamside disturbances included a cleared riparian zone and catchment dams immediately upstream.

Figure 3-1. Waurn Ponds Creek downstream of Anderson Road Crossing

Site 3- upstream Devon Road Crossing

Figure 3-2. Waurn Ponds Creek upstream of Devon Road Crossing

This survey reach extended from 10m to 60m upstream of Devon Road Crossing. The mean width and depth of the survey reach were 1.5 m and 0.15 m respectively. Hydraulic habitat comprised run (40%) and pool (60%). The streambed comprised clay (30%), silt (20%), fine sand (20%), coarse sand (20%) and boulder (10%). Instream habitat available as cover to aquatic fauna was dominated by aquatic vegetation (30%) and small areas of coarse substrate (5%). Streamside disturbances included a cleared riparian zone and a culvert.

Site 4- Cochran's Road Crossing

This reach extended from 30m upstream to 28m downstream of a ford approximately 500m downstream of the Cochran's Road crossing. The mean width and depth were 2m and 0.lm respectively. Hydraulic habitats comprised riffle (75%) and pool (25%). The streambed comprised largely coarse substrata including sand (40%) and boulder (30%). Instream habitat available as cover to aquatic fauna consisted of aquatic vegetation (75%) and coarse substrate (25%). Streamside disturbances included a mostly cleared riparian zone and an instream barrier.

Figure 3-3. Waurn Ponds Creek at Cochran's Road Crossing

Site 5- Jarvis Oval Bridge

Figure 3-4. Waurn Ponds Creek at Jarvis Oval Bridge

This reach extended from 55m downstream to approximately 21m upstream of Jarvis Oval Bridge. The mean width and depth of the survey were 1.5m and 0.4m respectively. Hydraulic habitats comprised riffle (30%), glide (20%) and pool (50%). The streambed comprised gravel (60%) silt (15%), clay (10%), sand (10%) and small areas of boulder (5%). Instream habitat available as cover to aquatic fauna consisted predominantly of aquatic vegetation (40%) and coarse substrate (30%). Streamside disturbances included a culvert and a partly cleared riparian zone.

Site 6- downstream Grovedale Road Crossing

This reach extended from approximately 350m to 400m downstream of Grovedale Road Crossing. The mean width and depth were 2m and 1 m respectively. Hydraulic habitats comprised run (85%), pool (10%) and riffle (5%). The streambed comprised silt (70%), clay (20%) and sand (10%). Aquatic vegetation dominated the available habitat for aquatic fauna (80%). Streamside disturbances included stream channelisation and incision.

Figure 3-5. Waurn Ponds Creek downstream of Grovedale Crossing

Site 7- Rotary Peace Park

Figure 3-6. Waurn Ponds Creek at Rotary Peace Park

This reach was located at Rotary Peace Park, and extended for a length of 60m. The mean width and depth were 1.5m and 0.7m respectively. Hydraulic habitats comprised run (85%), pool (10%) and riffle (5%). The streambed comprised predominantly sand (50%) and silt (38%). A variety of instream habitats were present including bank overhang (20%) and aquatic vegetation (20%). Streamside disturbances included stream channelisation and incision.

Site 8- Camdell Court

This reach was located off Camdell Court, and extended for 52m. The mean width and depth were 1.5m and 0.7m respectively. Hydraulic habitats comprised pool (70%) and run (30%). The streambed comprised gravel and sand (60%) and silt/clay (40%). Instream habitat comprised predominantly branches (30%), urban rubbish (20%) and vegetation overhang (15%). Streamside disturbances included stream channelisation and incision.

Figure 3-7. Waurn Ponds Creek at Camdell Court

Site 9- Barwon Heads Road

Figure 3-8. Waurn Ponds Creek downstream of Barwon Heads Road

This reach extended from the Barwon Heads Road Bridge to 43m downstream. The mean width and depth were 2m and 0.7m respectively. Hydraulic habitats comprised pool (60%), run (30%) and glide (10%). The streambed comprised predominantly fine sand, silt and clay (45% combined) and gravel (30%). Aquatic vegetation dominated the available habitat (40%). Streamside disturbances included two instream barriers (two fords with culverts), stream channelisation and incision.

3.3 FISH SURVEYS

3.3.1 Distribution and abundance

A total of 461 individual fish representing eight species (six native and two exotic) were recorded (Table 3-4). Species diversity and abundance tended to increase with increasing distance downstream from source. Species diversity ranged from two species in the upper reaches (site 2 and 3) to seven species at site 4. The number of individual fish recorded at each site also varied, ranging from 22 individuals at site 2 to 88 individuals at the most downstream site (site 9).

Common galaxias (Galaxias maculatus) were the most abundant species recorded, comprising 46% of the total catch, and were also one of the most widespread species, being recorded at all survey sites with the exception of two sites in the upper reaches (site 2 and 3). Yarra pygmy perch comprised 21% of the total catch, and were also one of the most widespread species, being recorded at the same six sites where common galaxias were recorded.

Short finned eel (Anguilla australis) were the most widespread species, being collected at all sites surveyed, although total abundance was low with only 23 individuals collected representing 5% of the total catch. Southern pygmy perch (Nannoperca australis) were collected at five sites and represented 13% of the total catch. Spotted galaxias (Galaxias truttaceus) were collected in low numbers from four sites. Flat-headed gudgeon (Philypnodon grandiceps) were recorded in low numbers from the most downstream site (site 9).

Exotic fish species were collected only in the upper catchment. Here water quality parameters displayed the highest temperature, electrical conductivity and turbidity (Table 3-2, 3-3). Eastern gambusia (Gambusia holbrooki) were collected in low numbers at three sites representing 10% of the total catch. One goldfish (Carassius auratus) was also collected.

Table 3-4. Total abundance of species collected in Waurn Ponds Creek.

	Site
Species	2	3	4	5	6	7	8	9	Total
Native teleost species
Anguilla australis	1	2	1	7	2	2	7	1	23
Philypnodon grandiceps								7	7
Galaxias maculatus			27	30	22	46	41	45	211
Galaxias truttaceus			1	3	1	12			17
Nannoperca australis			9	26	10		8	5	58
Nannoperca obscura			11	1	15	12	29	30	98
Exotic teleost species
Gambusia holbrooki	21	23	2						46
Carassius auratus			1						1
Total number of fish	22	25	52	67	50	72	85	88	461
Total number of fish species	2	2	7	5	5	4	4	5	8

Assemblage structure

Catch per unit effort (CPUE) varied between sites and tended to increase with increasing distance downstream from source (Figure 3-9). At site 2 and 3 located in the upper reaches of the catchment, the exotic eastern gambusia dominated fish assemblages with relative abundances of 95% and 92% respectively (Table 3-4). Short-finned eel were also recorded in low abundances (Table 3-4). In the mid- to lower reaches of the catchment species diversity and abundance increased compared to the upper reaches. Fish assemblages were typically dominated by common galaxias (41-68% relative abundance), and to a lesser extent Yarra pygmy perch and/or southern pygmy perch (Table 3-4, Figure 3-9). Spotted galaxias were also relatively abundant at site 7 in the lower reaches, comprising 17% of the fish fauna recorded at this site (Table 3-4, Figure 3-9). Exotic species were collected only from the three most upstream sites.

Figure 3-9. Catch Per Unit Effort (fish.hr^-1) for surveys at sites on Waurn Ponds Creek.

Abbreviations denote species as: Ca, Carassius auratus (goldfish); Gh; Gambusia holbrooki (Gambusia); No; Nannoperca obscura (Yarra pygmy perch); Na, Nannoperca australis (southern pygmy perch); Gt; Galaxias truttaceus (spotted galaxias); Gm, Galaxias maculatus (common galaxias); Pg, Philypnodon grandiceps (flat-headed gudgeon); Aa, Anguilla australis (short-finned eel).

3.3.3 Length-frequency distribution of Yarra pygmy perch

Figure 3-10. Length-frequency distribution for Yarra pygmy perch (n=68).

4. DISCUSSION

4.1 SPECIES DIVERSITY, DISTRIBUTION AND ABUNDANCE

The diversity of native fish species observed in the current study declined with increasing distance upstream. Although species diversity may decline as distance inland increases, owing to factors such as availability of suitable habitats and migratory ability (McDowall 1998), in Waurn Ponds Creek stream accessibility has been artificially reduced by human disturbance (e.g. instream barriers and stream channelisation). The relative dominance of native and exotic fish species varied throughout the study area. In general, fish communities in the upper reaches are dominated by exotic species, whereas in the mid- to lower reaches native species dominate fish assemblages. The diversity, distribution and abundance of fish observed in this study is likely to be a result of numerous anthropogenic factors in the catchment. These are discussed in more detail below.

4.2 SIGNIFICANT SPECIES

Yarra pygmy perch were found in the mid- to lower reaches of Waurn Ponds Creek and were abundant at five out of the six sites where it was collected (11-30 fish). The species is common in only a few places in Victoria but is generally not known to be abundant (most records of less than 10 fish) (Kuiter et aL 1996). An exception is in the nearby Thompson's Creek catchment where Zampatti (2001) collected 25-45 fish at four sites.

Yarra pygmy perch typically prefer aquatic vegetation (Koehn and O'Connor 1990) and have been collected in a range of water quality conditions, including low dissolved oxygen concentrations and high salinities (Zampatti 2001). Similarly, in Waurn Ponds Creek we collected Yarra pygmy perch predominantly in association with aquatic vegetation and in a range of water quality conditions.

Due to the wide distribution and abundance of Yarra pygmy perch in Waurn Ponds Creek it is likely that the Waurn Ponds Creek population is significant in southeastern Australia. Furthermore, the length frequency analysis presented here suggests that Yarra pygmy perch have successfully spawned and recruited into the adult population in recent years. Nevertheless, a number of factors may have a negative impact on Yarra pigmy perch and other native fish species in the catchment. These are discussed in section 4.3.

Although Australian mudfish has not been recorded in Waurn Ponds Creek, the species has been recorded from the Barwon River catchment (DNRE 2002) and we consider that under natural conditions the species may have potentially inhabited reaches of Waurn Ponds Creek. Australian mudfish is classified as 'endangered' in Victoria and is listed as a protected species under the Flora and Fauna Guarantee Act 1988. Due to the threatened status of Australian mudfish and the possibility that it may inhabit Waurn Ponds Creek, this species should be considered in the rehabilitation plan for Waurn Ponds Creek. Threatening processes that may impact on Australian mudfish are discussed below.

4.3 THREATENING PROCESSES IMPACTING ON AQUATIC FAUNA IN WAURN PONDS CREEK

Alteration of habitat through agricultural and grazing practices and, interaction (predation and competition) with exotic fish species have been previously recognised as key threatening processes to Yarra pygmy perch in southeastern Australia (Saddlier 1993). In addition, degradation of riparian vegetation, degradation of instream habitats (sedimentation, snag removal), reduced water quality (particularly dissolved oxygen and salt), alteration to the natural flow regime, modification to the stream channel (channelisation for flood mitigation) and instream barriers have also been identified as potentially impacting on populations of Yarra pygmy perch (Zampatti 2001) as well as other native freshwater fish (Koehn and O'Connor 1990).

• altered flow regimes;
• degradation of riparian (streamside) vegetation;
• degradation of instream habitat;
• channel modification and;
• instream barriers.

These impacts are discussed in more detail below.

Altered flow regimes

Alteration to the natural flow regimes of rivers and streams is considered a major threat to the health of waterways, aquatic flora and fauna communities and the maintenance of essential (instream) ecosystem processes (SAC 1992). For example, alterations to key components of the natural flow regime including flow magnitude, seasonality and variability are regarded as key factors responsible for the decline in distribution and abundance of many native freshwater fish species throughout Victoria (Koehn and O'Connor 1990). Alteration to natural flow regimes has been identified as a potentially threatening process under the Victorian Flora and Fauna Guarantee Act 1988 (SAC 1992) and has been nominated as a key threatening process under the Environment Protection and Biodiversity Conservation Act 1999.

Natural flow regimes may be altered through diversion of water for irrigation or stock and domestic use, catchment dams and on onstream storages, regulated releases from major headworks and catchment imperviousness in urbanised catchments or subcatchments, and stream channel modification.

In Waurn Ponds Creek, catchment dams, diversion of water, catchment imperviousness (lower reaches - Waurn Ponds), and channel modification may impact on aquatic fauna populations. Catchment dams may reduce flow magnitude and eliminate important components of the natural flow regime including freshes during summer months, which provide for critical ecological processes (eg. water quality maintenance over the low flow period and spawning keys for native fish species). Irrigation development may lead to reduced flow magnitude especially over the summer months, and an increase in the frequency and duration of low flow events. Although low flows are a natural component of the flow regime, an increase in their frequency and duration may represent a substantial increase in the risk to aquatic fauna populations through a reduction in water quality, available habitat and in food resources.

Catchment imperviousness in urbanised reaches of catchments results in alteration to natural run-off characteristics. Rapid rises and falls in water level and stream discharge are characteristic of these reaches and may physically displace aquatic fauna and habitat (e.g. large woody debris) and conversely, result in some biota becoming stranded when water levels recede rapidly. These effects are exacerbated in areas where the stream channel has been modified to reduce flooding (e.g. lower reaches of Waurn Ponds Creek). These channelised sections are designed to convey flood waters downstream and as a result areas of low water velocity in which aquatic biota can take refuge are rare. These stream modifications have also resulted in a reduction (possibly elimination) in the frequency of flood plain inundation. Flood plain habitats and off channel wetlands are important habitats for some species (e.g. Australian mudfish) and the inundation of floodplains is thought to provide a rich source of nutrients for instream ecological processes.

Although the degree of alteration to the natural flow regime was not investigated in this study, alteration the natural flows should be considered in the Waurn Ponds Creek rehabilitation plan.

Instream barriers

In the coastal drainages of south-eastern Australia up to 70% of native freshwater fish species may migrate between fresh and estuarine/marine waters at some stage during their life cycle (Harris 1984). A similar figure would be expected for Waurn Ponds Creek. In Victoria however, nearly 2200 artificial structures in rivers and streams have been documented as potentially restricting fish movement (McGuckin and Bennett 1999). These structures include barriers such as stream gauging stations, on-stream dams and weirs, fords and culverts.

Unimpeded passage of fish throughout streams is crucial for spawning migrations, recolonizations, general movement and habitat selection (Koehn and O'Connor 1990) and is therefore considered important in sustaining biodiversity in stream fish assemblages. Fish movement is protected by Victoria legislation through a number of Acts (Water Act 1999, Fisheries Act 1995, Flora and Fauna Guarantee Act 1988, Conservation, Forests and Lands 1987) and "...the prevention of passage of aquatic biota as a result of instream structures..." is identified as a potentially threatening process (Flora and Fauna Guarantee Act 1988).

Numerous barriers (fords, culverts and drop structures) which likely restrict movement (migrations and localised movements) of fish were identified on Waurn Ponds Creek. The restricted distribution of native fish species observed in this study as well as the absence of many native migratory species expected to occur in the catchment may be a result of numerous barriers to fish movement. Unimpeded passage for native freshwater fish should be considered in the Waurn Ponds Creek rehabilitation plan.

Degradation of riparian vegetation

Riparian (streamside) vegetation provides important links between the aquatic and terrestrial environments. For example, streamside vegetation acts as a buffer zone for the instream environment and filters sediment input as well as other diffuse source pollutants e.g. nutrients, pesticides etc. Streamside vegetation also contributes to instream habitat (leaf litter, stick fall) and provides an important source of terrestrial invertebrate prey for aquatic biota.

Waurn Ponds catchment has been extensively cleared for agriculture in the upper reaches and is urbanised in the lower reaches. Much of the streamside vegetation has been removed and critical ecological process (see above) reliant on it have been disrupted. Some revegetation has occurred in the lower reaches of the catchment and already appears to be benefiting aquatic biota. For example, of the few spotted galaxias collected in fish surveys in Waurn Ponds Creek, most individuals were collected from a revegetated reach from habitat created as a result of this revegetation (overhanging leaves and woody debris from recent stick fall).

Intact and healthy riparian vegetation is critical for instream biota and should be considered in the Waurn Ponds Creek rehabilitation plan.

Degradation of instream habitat

The complexity and diversity of instream habitat is critical for supporting a diversity of aquatic fauna. A variety of processes including those discussed above interact to reduce habitat complexity and availability in Waurn Ponds Creek. As a result of clearing of catchment and riparian vegetation, large quantities of sediment have accumulated in Waurn Ponds Creek channel. This sediment smothers complex habitats (cobbles, rocks, large woody debris, aquatic vegetation) and reduces the habitat available for aquatic fauna. Stream channelisation has also reduced instream habitat complexity.

This study has identified numerous factors impacting on environmental condition and more specifically on aquatic fauna assemblages in the catchment. Consideration of these issues in the rehabilitation plan for Waurn Ponds Creek will help to maintain and restore aquatic biodiversity in Waurn Ponds Creek, including populations of the threatened Yarra pygmy perch. Based on the information collected in this study and the discussion of issues presented above, we recommended the following priority issues for consideration in the Waurn Ponds rehabilitation plan:

• degradation of instream habitat;
• barriers to fish migration;
• degradation of riparian (streamside) vegetation;
• changes to stream channels and;
• changes to natural flow patterns.

5. ACKNOWLEDGEMENTS

6. REFERENCES