Cloverdale Project M-CV-00028

Baseline Aquatic Biology and Water Quality Study including Tiger Gully, Ludlow River and Capel River

prepared for

by

Wetland Research & Management

Cloverdale Project M-CV-00028

Baseline Aquatic Biology and Water Quality Study including Tiger Gully, Ludlow River and Capel River

Prepared for:

Iluka Resources Limited Level 23, 140 St Georges Terrace, Perth WA 6000 GPO Box U1988 Perth WA Ph (61 8) 9360 4700

By:

Wetland Research & Management 28 William Street, Glen Forrest, WA 6071, Australia Ph (61 8) 9298 9807, Fax (61 8) 9380 1029, e-mail: awstorey@ cyllene.uwa.edu.au

Draft Report April 2006

Frontispiece: (clockwise from main picture) Ludlow River, 1.2 km downstream from the Cloverdale project area; male koonac Cherax plebejus; nightfish Bostockia porosa.

ii

Study Team Management: Sue Creagh Field Work: Sue Creagh & Jess Lynas Macroinvertebrate Identification: Lisa Chandler & Sue Creagh Data analysis: Sue Creagh & Andrew Storey Report: Sue Creagh

Acknowledgements This project was undertaken by Wetland Research & Management (WRM) for Iluka Resources Limited. WRM would like to acknowledge Dr Don Edward (UWA) for assistance with Chironomidae taxonomy and Dr Mark Harvey (WAM) for Acarina taxonomy and Dr Rob Davis (Western Wildlife) for tadpole identifications. The maps of the study area were provided by Iluka Resources Limited. Shannon Jones (Iluka) is thanked for constructive criticism on the draft report and for her efficient overall management of this project on behalf of Iluka. The authors are grateful to Craig and Tom Hutton, Garry Bibby, Jan McKechnie and to the Norton, Armstrong, Weir and Whiteland families, who readily granted access to their pastoral properties. Tom Hutton is especially acknowledged for his assistance in the field and for providing valuable information on the fauna and ecology of the Capel River. Peter Nelson and Denise Barnard (both Iluka) are thanked for kindly providing assistance on-site and Rae McPherson (Capel LCDC) provided water quality monitoring data for Capel River.

Recommended Reference Format WRM (2005). Cloverdale Project M-CV-00028: Baseline Aquatic Biology and Water Quality Study including Tiger Gully, Ludlow River and Capel River. Unpublished report by Wetland Research & Management to Iluka Resources Ltd. April 2006.

Disclaimer This document was based on the best information available at the time of writing. While Wetland Research & Management (WRM) has attempted to ensure that all information contained within this document is accurate, WRM does not warrant or assume any legal liability or responsibility to any third party for the accuracy, completeness, or usefulness of any information supplied. The views and opinions expressed within are those of WRM and do not necessarily represent Iluka Resources Limited policy. No part of this publication may be reproduced in any form, stored in any retrieval system or transmitted by any means electronic, mechanical, photocopying, recording or otherwise, without the prior written permission of Iluka Resources Limited and WRM.

iii Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

CONTENTS SUMMARY...... vi 1. INTRODUCTION ...... 1 2. METHODS...... 1 2.1 Study Area ...... 1 2.2 Sites and Sampling Design ...... 2 2.3 Riverine Condition Assessment...... 5 2.4 Physico-chemistry...... 5 2.5 Macroinvertebrate Fauna ...... 6 2.6 Fish and Crayfish...... 7 2.7 Frogs and Waterbirds...... 7 2.8 Data Analysis...... 7 2.8.1 Univariate Analysis...... 7 2.8.2 Multivariate Analysis ...... 7 3. RESULTS AND DISCUSSION...... 9 3.1 Riparian and In-stream Habitat Condition...... 9 3.2 Physico-chemistry...... 12 3.2.1 Patterns in Physico-chemical Data...... 15 3.2.2 Comparisons with Known Water Quality Data for Other Systems in the Area ...... 18 3.3 Macroinvertebrates ...... 18 3.3.1 Taxonomy and Species Richness...... 18 3.3.2 Conservation Significance of Macroinvertebrates...... 20 3.3.3 Functional Feeding Groups...... 22 3.3.4 Patterns in Macroinvertebrate Community Structure...... 23 3.3.5 Comparisons with Other Macroinvertebrate Studies...... 28 3.4 Fish and Crayfish...... 28 3.4.1 Conservation Significance of Fish Fauna...... 29 3.5 Frogs ...... 30 3.6 Waterbirds...... 31 4 CONCLUSIONS...... 32 4.1 Ecological Values ...... 32 4.2 Potential Threats from Mine Activities...... 33 4.3 Recommendations...... 34 5 REFERENCES ...... 36 APPENDICES...... 39 Appendix 1. Summary of physico-chemical data from each site...... 40 Appendix 2. ANZECC/ARMCANZ (2000a) guideline levels ...... 42 Appendix 3. PCA results for physico-chemical variables ...... 45 Appendix 4. Macroinvertebrate taxa occurrences across sites...... 49

______Wetland Research & Management ______iv Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

List of Figures, Tables and Plates Figures Figure 1. Location of aquatic ecosystem survey sites...... 3 Figure 2. Changes in salinity (as Econd.) within the Ludlow /Tiger Gully catchment...... 13 Figure 3. Iluka water quality data showing longer-term changes in salinity, sulphate and pH at Site 1 (YDP)...... 14 Figure 4. Two-dimensional PCA ordination plots of physico-chemical variables...... 17 Figure 5. Species richness across sites surveyed...... 19 Figure 6. Proportional occurrence of functional feeding groups of macroinvertebrates collected...... 22 Figure 7. Cluster dendrogram using macroinvertebrate log10 abundance data, indicating four main groupings...... 24 Figure 8. MDS ordinations using macroinvertebrate log10 abundance data for all sites...... 24 Figure 9. MDS ordinations using Ludlow/Tiger Gully macroinvertebrate log10 abundance data...... 25 Figure 10. Bubbles plots superimposed on the MDS ordination, showing examples of the variation in abundance classes between sites for four macroinvertebrate taxa...... 27 Figure 11. Bubble plots superimposed on the MDS ordination, showing examples of physico-chemical variables that correlated best to patterns in macroinvertebrate abundance classes...... 27

Tables Table 1. Co-ordinates of sites sampled for aquatic fauna...... 4 Table 2. Physico-chemical parameters measured at each site...... 5 Table 3. Summary of riverine condition...... 11 Table 4. Eigen values, percent of variance and cumulative percent of variance from PCA...... 17 Table 5. Summary of invertebrate fauna collected ...... 19 Table 6. Sites ranked by number of SW endemics...... 21 Table 7. Matrix of association values indicating percent pairwise similarity amongst sites, based on macroinvertebrate abundance data...... 26 Table 8. Fish and Crayfish species recorded ...... 29 Table 9. Frog species listed by Bamford (2001) as present in the Ludlow area...... 31 Table 10. Waterbird species and numbers recorded...... 32

Plates Plate 1. Site 1 (lot 3739), view west toward confluence with Tiger Gully...... 9 Plate 2. Site 3, Ludlow River on lot 2, uppermost control site...... 9 Plate 3. Site 2 (YWLRU), Ludlow River on lot 1392. Immediately upstream from Ludlow diversion point...... 10 Plate 4. Site 4 (TGD; lot 1188), Iluka’s Tiger Gully rehabilitation area downstream from Yoganup West ...... 10 Plate 5. Site 5, Tiger Gully at confluence with old Ludlow River channel (lot 1392) ...... 10 Plate 6. Site 6, Ludlow River, lot 2015; conservation area immediately east of Warns Road ...... 10 Plate 7. Site 8, Ludlow River conservation area bordering lots 3229 & 2012 ...... 10 Plate 8. Site 7, Ludlow River conservation area. Most downstream site on lots 3229 & 2012...... 10 Plate 9. Site 9 on the Capel River (lot 3299), January 2006...... 11

______Wetland Research & Management ______v Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

SUMMARY Iluka Resources Limited propose to mine mineral sands at Cloverdale, near Capel in the south-west of Western Australia. The Cloverdale project area is located on (largely) cleared agricultural lands east of Bussell Highway, between Warns Road and the existing Yoganup / Yoganup West mines. The Ludlow River and Tiger Gully traverse the project area but the channels will not be mined. The development has the potential to adversely affect the aquatic ecosystem of the downstream conservation area along the Ludlow River. The goal for Iluka is to maintain the existing biodiversity of this seasonal system. To that end, Wetland Research & Management (WRM) sampled the aquatic fauna and physico-chemistry at eight sites along the Ludlow River/ Tiger Gully system in November 2005 and one site on the Capel River in January 2006. Sampling represented a baseline study to help establish conditions prior to commencement of mining, against which future changes (if any) may be assessed. The catchments are already undergoing secondary salinisation as a result of clearing for agriculture and there are also areas of potential acid sulphate soil risk. The specific aims of the project were to identify the resident fauna to the lowest possible taxonomic level, compare biodiversity and invertebrate assemblages amongst sites and relate differences in fauna to differences in water quality. Water temperature, dissolved oxygen, pH, salinity and Secchi depth were measured in situ. Water samples were collected for laboratory analyses of nutrients, alkalinity, water hardness, soluble solids, petroleum hydrocarbons and metals. Aquatic macroinvertebrates and tadpoles were collected using qualitative sweeps. Sampling for fish and crayfish was conducted using an electrofisher, baited box traps and by direct observation. Broad, qualitative assessments of riparian habitat condition were made on the basis of dominant plant species and erosional characteristics. Opportunistic surveys of waterbirds were also made.

Riparian Habitat Condition All sub-catchments were considered extremely degraded due to historic pastoral practices and unrestricted livestock access to the natural waterbodies. Terrestrial native understorey vegetation was, at best, sparse; the understorey dominated by pasture grasses (e.g. Kikuyu) and weeds, which for the most part was providing a degree of protection against excessive bank erosion. Along the river, overstorey vegetation ranged from open to moderately dense mixed woodlands of marri (Corymbia calophylla), blackbutt (Eucalyptus patens) and peppermints (Agonis flexuosa) with some river red gums (E. rudis) and paperbarks (Melaleuca rhaphiophylla). Outside of the conservation wetland area, most trees observed were mature, with little evidence of recruitment. An environmental rating of ‘moderate’ was assigned to the Ludlow conservation area and Capel River sites, but all other sites were rated as ‘poor’ to ‘very poor’ on the basis of health and percentage cover of remnant vegetation and the degree of bank/bed erosion. The regional ecological value of the remnant riparian vegetation and in particular that along the Capel River and within the Ludlow River conservation category wetland area, was considered to be high, given the extensive (historic) clearing of native vegetation for agriculture. The remnant vegetation is likely to provide at least some of the energy (‘food source’) that drives many aquatic processes (e.g. food webs) as well as providing food, shade and shelter for both terrestrial and aquatic fauna. Where ever possible, the trees and few remaining perennial shrubs should be conserved to support food webs and to help stabilise the river banks against further erosion.

Physico-chemistry Values for all of the physico-chemical variables measured were well within ANZECC/ARMCANZ (2000a) guideline limits for livestock drinking water and most were within the ranges expected for ‘slightly to moderately disturbed’ ecosystems (ANZECC/ARMCANZ 2000a). The exceptions included salinity (ECond range 411 - 819 JS/cm), which exceeded the maximum trigger value of 300 JS/cm (ANZECC/ ARMCANZ 2000a) for aquatic ecosystems in south-west rivers. There was a slight gradient in salinity in the Ludlow/Tiger Gully system, with values tending to increase longitudinally down the catchment. The composition of major ions in the Ludlow catchment was typically dominated + 2+ + 2+ - 2- - by sodium and chloride (Na >Mg >K > Ca : Cl>SO4 >HCO3 ), however, the concentration of

______Wetland Research & Management ______vi Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06 calcium and sulphate ions was much greater near the Yoganup discharge point (YDP Site 1) in the Tiger Gully catchment upstream of the Cloverdale project area. Salinity (& associated ion concentrations), water hardness and alkalinity all tended to be greatest near YDP. There are currently no guideline recommendations for maximum levels of water hardness, alkalinity, sulphate or calcium for the protection of aquatic ecosystems and levels were within current guidelines for livestock drinking water. Data collected by Iluka at YDP over the period 1999-2005 show what may be a slight upward trend in salinity and sulphate levels, post November 2003, and a downward trend in pH during 2005. The changes are minor and within the limits imposed by the current Yoganup/Yoganup West licence (ECond. 1,250 µS/cm & pH 5.5 - 9.0 pH). Though there are no comprehensive records for Tiger Gully prior to mining, levels would still appear to be within the seasonal ‘background’ range known for the Ludlow River and nearby Capel River. However, levels should continue to be monitored to determine if they do indeed represent actual trends and not just short-term fluctuations. Zinc (Zn) exceeded aquatic ecosystem guidelines at all sites, but in particular at YDP. Even when water hardness at YDP was taken into account, Zn levels were more than twice the recommended maxima. The high level of zinc at all sites was taken as indicative of the geology of the area, though agricultural fertilizers (e.g. zinc sulphate) may also be a source. Mining however, may have resulted in the relatively greater concentrations at Site 1, but no baseline data were available for comparison. Nickel was also elevated at YDP compared to all other sites. Aluminium (Al) and cadmium (Cd) concentrations at control Site 3 on the Ludlow River were slightly above aquatic ecosystem guideline limits and up to ten times greater than at all other sites. These elevated levels were not considered due to mining as this site lay outside the influence of the Yoganup and Yoganup West mine sites. It must also be noted that detection limits for Cr and Cu were between two to five times greater than aquatic ecosystem guidelines.

While NO3 concentrations at two sites (control Site 3 & Ludlow conservation area Site 6) approached ANZECC/ARMCANZ (2000a) ecosystem guideline values, other nutrient readings were unexpectedly low.

Aquatic Fauna M acroinvertebrates In total, 111 macroinvertebrate taxa were recorded from the Ludlow River and Tiger Gully in November 2005, with an additional 24 taxa recorded from the Capel River in January 2006. This represented a moderately diverse fauna for seasonal waters and perennial waters, respectively. Community composition was similar to other disturbed seasonal and perennial watercourses on the coastal plain. The fauna was dominated by insects (74%) with at least 62 species representing 18 families. Chironomidae (non-biting midges) constituted 24% of all insect taxa. Crustacea (including freshwater crayfish) constituted only 9% of the total fauna. Taxa normally associated with less disturbed ecosystems were rare or absent (e.g. Ephemeroptera, Odonata and many of the Trichoptera) in the Ludlow, but more abundant in the Capel River. Most taxa were considered common, i.e. abundant and widespread throughout WA and occurring in other states/territories. Twenty-five taxa (18%) recorded were south-west endemics, but only one, the freshwater mussel Westralunio carteri, were considered restricted within the south-west. This mussel, which was only recorded from the Capel River, is currently listed as a Priority 4 species under CALM’s Wildlife Conservation (Specially Protected Fauna) Notice 2005 and as ‘vulnerable’ under the IUCN Red List of Threatened Species (2004). These listings indicate that whilst not currently threatened, mussel populations are fragmented and in need of monitoring. Population decline has been reported in many areas throughout the south-west and is likely related to secondary salinisation and heavy sedimentation/siltation of river beds and pools.

______Wetland Research & Management ______vii Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Other south-western endemics included two native freshwater crayfish (marron, gilgies & koonacs) and three amphipods (Austrochiltonia subtenius, Perthia branchialis & P. acutitleson). Gilgies (Cherax quinquecarinatus) were present throughout the system, but koonacs (C. plebejus) were only recorded either side of Warns Road. Marron were only recorded from the perennial Capel River. Amphipods were only present at the most downstream sites within the Ludlow River conservation category wetland area and in the Capel River. The Capel River was found to support considerably more endemic species than the Ludlow system, most likely due to the perennial flow regime coupled with remnant overstorey vegetation. All insects recorded, were considered to be ‘temporary’ residents with highly mobile adult phases that would allow them to avoid adverse environmental conditions and reinvade from nearby permanent waterbodies, once conditions improved. Species such mussels and marron, are more at risk as they are permanent residents with less capacity to escape adverse conditions. Species diversity at individual sites ranged from 25 (Site 3, upper Ludlow River control) to 69 (Site 7, conservation area along Ludlow River). No one taxa could be viewed as truly indicative of control or of exposed (potential impact) groups or of an individual site. However, Site 3 did differ from all other sites due to the absence of trichopteran (caddis-flies) and hemipteran (true bugs) species and the presence of very few dytiscids (aquatic beetles). Despite this and despite differences in water chemistry between sites, multivariate analyses (PRIMER) showed there were no significant differences in assemblages or taxa richness, with only low to moderate between-site variation. PRIMER analyses indicated the inter-site variability in assemblage composition that did exist was possibly influenced by local variations/between-site differences in Al, Cd, Ni, Mn, Ba, NH3, pH, salinity (TDS, ECond. & associated concentrations in K+ and Na+), channel width and channel depth. Fish Four species of fish (western minnows, nightfish, western pygmy perch & mosquitofish) were collected from the study area and local landowner Tom Hutton, reported an additional three species as present in the Capel River – cobbler, redfin perch and pouched lampreys. Redfin perch and mosquitofish are both introduced species. Mosquitofish were particularly abundant at sites downstream of the Cloverdale Project Area. There were no noticeable differences in abundance of native species between control and exposed sites in the Ludlow catchment. All native species are common and widespread throughout the south-west. Though not recorded during the current study, the swan river goby may also be present within the project area it has previously been reported in the lower Ludlow River. Tadpoles Tadpoles of only one species, the slender tree frog (Littoria adelaidensis), were collected in low numbers from control Site 3 on the Ludlow River. This species is widespread and abundant throughout the south-west. Turtles Turtles were not specifically targeted during the current surveys and none were sighted. However, western long-necked turtles (Chelodina oblonga) are known to be common in the Capel River (Tom Hutton, Capel, pers. comm.). C. oblonga is widespread and abundant throughout the south-west. Waterbirds Waterbird species were observed feeding along the Ludlow River/Tiger Gully and included Pacific black duck, grey teal, Australian wood duck, straw-necked ibis, white-necked heron, white faced heron, black-fronted dotterel. Flocks of straw-necked ibis were also observed in paddocks adjacent to the river. The narrow river channel and lack of a large, open water body means the Capel and Ludlow rivers are unlikely to support an abundance of waterbirds. Loss of fringing vegetation has likely reduced suitable nesting sites across the floodplain and nearby wetlands. All species reported are common. None are listed under JAMBA/CAMBA treaties.

______Wetland Research & Management ______viii Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Water Rats Though not recorded during the current study, native water rats Hydromys chrysogaster are known to be present along the Capel River within the study area (Tom Hutton, Capel, pers. comm.). Water rats are common around coastal Australia and New Guinea, occurring in a wide range of coastal, brackish and freshwater environments.

Conclusions The overall ecological health of the study area was assessed as ‘poor’ with unstable channels (including excessive erosion at some sites) supporting only a poor to moderate diversity of habitats and aquatic fauna characteristic of other disturbed southwest rivers in rural areas. However, along the Capel River and in the conservation area along the Ludlow River downstream of the proposed development, remnant overstorey vegetation, good tree recruitment and greater macroinvertebrate biodiversity confer a relatively high conservation value on these sections. As the Ludlow system is naturally seasonal but with higher than historic winter flows and rising water tables due to clearing, projected localised groundwater drawdown is not expected to adversely effect existing aquatic ecosystems. Any potential changes in surface flow regime posed by mine activities are not expected to affect the vegetation of the Ludlow River conservation area which was considered more reliant on groundwater. The extent to which vegetation along the Capel River is reliant on groundwater could not be ascertained within the scope of this study. Any reduction in surface flows is likely to be compensated by overland paddock flows. A substantial reduction in river flows coupled with significant groundwater drawdown and lowered soil moisture would need to occur before downstream ecosystems were adversely impacted. However it must be noted that perennial flows in the Capel River would already appear under threat from excessive abstraction for summer irrigation for agriculture. While current mine operations at Yoganup and Yoganup West have had no apparent adverse effect on aquatic fauna communities, there is the suspicion that mining may be affecting water quality due to increased salinity, water hardness, alkalinity, sulphate, Zn and Ni. For the metals at least, this effect would appear to be highly localised. Other possible influencing factors which must also be considered are the unique broad, flat wetland and local hydrology / geology at YDP resulting in increased salts, water hardness, Ni and Zn levels in the shallow waters of the drainage line.

Recommendations A number of recommendations are made in regard to on-going monitoring, in particular water quality monitoring in view of elevated parameters at YDP. Any water analyses conducted by external laboratories must provide detection limits for analytes (e.g. Cu, Cr, benzene & xylene) that meet ANZECC/ARMCANZ (2000a) trigger values.

______Wetland Research & Management ______ix Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

1. INTRODUCTION This report documents the findings of a baseline aquatic ecosystem survey of the Ludlow River and Tiger Gully. The survey was conducted as part of an Environmental Impact Assessment for Iluka Resources Limited proposed mineral sands mining development, ‘Cloverdale’, near Capel in the south- west of Western Australia. Ecosystems in the vicinity of the existing Yoganup West mine were also included in the current survey as upstream control sites for the Cloverdale project. Waterbodies within the study area were surveyed using established CALM protocols. Surveys were in accordance with The Environmental Protection Act 1986 (Part IV), the EPA Guidance Statement No. 10 (January 2003) Guidance Statement for level of assessment for proposals affecting natural areas within the System 6 region and Swan Coastal Plain portion of the System 1 region and the EPA Guidance Statement 33 (June 2005) Environmental Guidance for Planning and Development1.

The aim of the surveys was to identify riverine systems potentially at risk, determine the aquatic fauna present at each site (aquatic macroinvertebrates, fish, tadpoles and waterbirds), determine conservation significance of the fauna and (for Cloverdale) establish baseline conditions against which future mine- related changes may be assessed.

2. METHODS 2.1 Study Area The area of interest for this study was the mid and upper reaches of the seasonally-flowing Ludlow River (east of Bussell Highway and west of the Whicher Scarp), including the Tiger Gully tributary. Both the river and gully flow through the Cloverdale project area east of Warns Road, Capel (Figure 1). Most of the Ludlow River between Warns Road and downstream Capel-Tutunup Road is listed by the Environmental Protection Authority (EPA) as conservation category wetland. The Ludlow River ultimately flows into the ecologically important Vasse-Wonnerup Estuary, approximately 11 km to the west. To the north, the perennial Capel River flows close to, but outside, the mine lease boundary. Historically, the Capel River also flowed into the Vasse-Wonnerup wetlands, but now flows directly to the sea following the construction of an artificial river mouth in 1880 (White & Comer 1999). Sandbar formation at the river mouth means that today the river is only connected to the ocean during winter and only then during high flow.

The Ludlow and Capel rivers are both located within the Geographe Bay catchment, which encompasses the shires of Busselton and Capel, 250 km south of Perth. Topography is typically low relief and poorly drained with many areas water-logged in winter. There has been extensive drain construction and clearing for agriculture within the catchment, the majority of which was conducted prior to the 1980s. Loss of riparian vegetation, gullying and bank erosion are common problems along both natural and artificial drainage lines. Loss of vegetation leads to loss of biodiversity and has had widespread ramifications downstream; e.g. increased nutrient, salt and sediment loads and increased surface runoff resulting in flooding and channel erosion. Historically, riparian zones would have been wide and densely vegetated with winter-wet depressions and swamps on floodplains during winter. The Geographe Catchment Council (GeoCatch) was established in 1997 in response to community concern over the declining health of waterways within the catchment. GeoCatch implemented an integrated catchment management approach which has included fencing riparian zones, streamlining and revegetation of creeks and rivers. GeoCatch has already issued a River Action Plan for the Capel River (White & Comer 1999) and is in the process of formulating one for the Ludlow River.

The channels of Ludlow River and Tiger Gully between Cloverdale and Warns Road are already significantly altered from the natural condition. Most recently, the Ludlow River has been diverted at

1 http://www.epa.wa.gov.au/docs/GS33/2060_GS33_PartB.pdf

______Wetland Research & Management ______1 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Site 2 (YWLRU) into the straightened channel of Tiger Gully to accommodate the existing Yoganup West mine. The Yoganup West discharge point is shown on Figure 1. Mining at Yoganup West commenced in 2004. Iluka has already commenced streamlining and rehabilitation of Tiger Gully below Yoganup West. In the past, this reach of the Gully had been straightened to increase drainage for agriculture. To the east of Yoganup West lies another of Iluka’s project areas, Yoganup. Mining at Yoganup commenced almost 50 years ago. Dewatering discharge from Yoganup contributes to the seasonal flows in both the gully and river. The old Ludlow River channel still conveys local overland runoff, joining Tiger Gully at the old confluence.

Under the Cloverdale proposal, there will be no mining through the Tiger Gully channel or the Ludlow River channel below the confluence, though there are a few shallow agricultural drains that traverse the mine path. No new solar drying dams will be located at Cloverdale, instead existing facilities will be utilised and all mine drainage will be internal.

While flow in the Ludlow River and Tiger Gully is seasonal, there are natural permanent and semi- permanent pools within the river that are maintained over summer by groundwater. Known pools are located in the upper catchment above the Yoganup West project area. There are no known natural permanent pools within the project area (Shannon Jones & Stuart Simmons, Iluka, pers. comm.). Headwaters of the Capel River are regulated by a large, private dam in the upper catchment near Donnybrook, some 20 km east of the project area, and by numerous smaller farm dams which supply water for irrigation. Though naturally perennial, the Capel River has ceased to flow on occasion in recent years due to excessive pumping for summer irrigation (Tom Hutton, Capel, pers. comm.). There is anecdotal evidence that while the river used to flood every year, it now only floods once in every 5 - 10 years – a response to both increased rates of surface water abstraction and decreasing rainfall (Tom Hutton, Capel, pers. comm.).

2.2 Sites and Sampling Design Following consultation with Iluka staff and a walkover of the area in September 2005, eight sites on the Ludlow system and one on the Capel River (Figure 1 & Table 1) were chosen to document current ecological values of the waterbodies potentially at risk.

Sampling of the Ludlow system was conducted during late spring 2005 (8th - 15th November). In order to differentiate potential effects of the existing Yoganup West project from those of the proposed Cloverdale project, sites were located along the Ludlow River, upstream and downstream of each project. This sampling design was broadly based on two complementary approaches to bio-monitoring: (i) using a classic, hypothesis-testing framework by way of the Before-After-Control-Impact class of design (BACI), and (ii) multivariate analysis of changes in community structure. Both approaches rely upon sampling replicate sites to characterise spatial variability in the parameters being measured (i.e. species richness, assemblage composition, water quality parameters) and to provide statistical power (viz. the ability to statistically detect differences/affects if they exist). The basis of the BACI design is a comparison through time, before and after mine development, between fauna community responses from control and exposed (potential impact) sites (Table 1) using classic analysis of variance (ANOVA) and multivariate test methods. If disturbances from mining were evident in future, this would be detected by departure in biological responses between control and exposed sites compared to the same responses measured (i) from the same sites before development and (ii) in control sites throughout the monitoring period.

______Wetland Research & Management ______2 Lot 10 Lot 1057 Lot 3350 Lot 3813 Lot 2382 Lot 11 Lot 3246 Lot 3141 Lot 31 Lot 100 Lot 2374 Lot 2034 R2850 Lot 1 Lot 3097 Lot 3775 G Lot 30 Lot 3841 Lot 101 o o Lot 4210 Lot 12 d Lot 20 w Lot 2253 o VCL C o ap Lot 1124 d e Lot 21 Lot 1092 R Lot 2432 l R o i Lot 4571 a ve Lot 22 Lot 4572 Lot 11 d r Lot 1124 Lot 3299 Road ntation Lot 3830 Wa Lot 4221 Lot 75 r n Lot 3840 Lot 3834 Lot 440 Lot 23 s Lot 12 Lot 3096

Lot 3809 R o Lot 1 Lot 504 a d Lot 2 Lot 2259 Lot 3838 Lot 2393 Lot 3782 R12211 7 Lot 1328 Lot 4218

(! Lot 4196 d

a Lot 3229 Lot 3779 o Lot 2393

R e

l

p a Lot 1453 u d

n r u

t e Lot 4216

u v

T Lot 3720

VCL o

- 8 l Lot 2012 Lot 4227 Lot 2052 l C e (!

p Ludlow River a C (!6 Lot 1442 VCL 5 Lot 1468 Lot 2014 (! Lot 2013 Lot 2015 Lot 1 4 (TGD) Lot 2066 (! Lot 3828 Lot 1188 Lot 976 Lot 1180 est Lot 2011 up W Lot 2018 Lot 2017 Lot 2016 an Lot 3740 Lot 3821 Yog

C 1 (YDP) l Lot 3739 Lot 2733 o

v (!

Lot 2021 e

Lot 3756 r

d

a

l

e

Lot 4075 Lot 2 Lot 1392 R

Lot 2019 Lot 2020 o Lot 2007 Lot 2 2 (YWLRU) a (! d Lot 4736 Lot 4075 p u Lot 2023 Lot 4411 n Lot 2006 a Lot 1 Lot 2023 Lot 68 g Loc 2008 Lot 2022 Lot 975 Lot 3835 o Lot 2024 Y Lot 2006

Lot 2024 Lot 368 Lot 3820 Lot 702 Lot 2266 Concentrator Lot 2 Lot 1326 Lot 3201 Lot 236 Lot 3201 Lot 1326 Lot 1

Lot 1 Lot 3087 Lot 3794 Lot 2003 Lot 638 Lot 1 Lot 1 Loc 2007 Lot 2 3 Lot 2003 (! t Lot 3203 Lot 3202 Lot 2 Lot 2716 Lot 2716 unu p R Legend

Drains CLOVERDALE Ludlow River Final LOCATION OF No Mining Area Project Area ± AQUATIC ILUKA HM Resource ORIG: SJones ECOSYSTEM Meters Roads 0137.5 275 550 825 1,100 DRAWN: SJones SURVEY SITES Cadastre SCALE: 1:35,000 MGA Coordinates, GDA94 (! Survey Sites DATE:24 Feb 2006 DWG No: APJCVAquSurvSts060224 A4 FIGURE: 1 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Table 1. Co-ordinates of aquatic fauna sampling sites in the Ludlow/Tiger Gully system and Capel River.

Project Aquatic Location Lot No. UTM (WSG84) Zone 50 Site No. Easting (m) Northing (m) Cloverdale & 1 Tiger Gully near Yoganup discharge point 3739 370616 6277731 Yoganup West (YDP). 2 Upper Ludlow River immediately upstream of 1392 369294 6277245 Ludlow diversion (Iluka WQ sampling point YWLRU) but downstream of dairy. 3 Upper Ludlow River, most upstream control 2 369568 6274738 site. 4 Tiger Gully rehabilitation area, downstream 1188 369129 6278532 of Ludlow diversion (Iluka WQ sampling point TGD). 5 Confluence of old Ludlow River and Tiger 1180 368529 6278800 Gully. 6 Ludlow River conservation wetland area 2015 368095 6279106 downstream of mine resource. Cloverdale 7 Ludlow River conservation wetland area 3229 / 2012 366609 6280081 downstream of mine resource. 8 Ludlow River conservation wetland area 3229 / 2012 367125 6279345 down-stream of mine resource. 9 Capel River, adjacent to northern boundary 3299 371100 6281779 of mine resource.

Sites were selected to maximise the biodiversity recorded, provide replicates satisfying the above design and provide a geographical spread with a range of different physical characteristics and types of remnant vegetation communities (including conservation category wetland reaches along the Ludlow River). Final locations of some sites were also dependent on access through private properties. The inclusion of Site 4 will also help establish the success of Iluka’s rehabilitation/streamlining programme in re-creating sustainable aquatic ecosystems in the mid-reaches of the Ludlow River diverted for Yoganup West.

Agricultural drains that pass through the project area to the north of Tiger Gully and to the west of the old Ludlow River channel were visually assessed and considered of limited ecological significance due to their artificial construction, poor bank stability, lack of in-stream habitat, seasonal flow and the degraded state of the floodway and verge vegetation. They were considered unlikely to support fauna of high conservation value and were therefore not included in the current sampling.

Though Iluka expects the area of groundwater drawdown will be highly localised, a targeted assessment of the Capel River was conducted at one site in order to document the existing ecological values of the river reach immediately adjacent to the northern boundary of the mine lease (i.e. on lot 3299). The site was sampled January 2006. Summer sampling was considered the most suitable approach to monitoring this perennial river, as any drawdown effects will more likely manifest when flows are low and fauna concentrated into ‘refuge’ pools. Any potential water quality problems are also likely to be exacerbated over summer.

Four EPP and two resource enhancement category wetlands are located over 2 km NW of the resource, well outside the predicted zone of drawdown. As such, it is highly unlikely that these wetlands will be directly influenced by mining activities such as dewatering and therefore they were not

______Wetland Research & Management ______4 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06 assessed. The nearest Ramsar2 listed wetlands are McCarly’s Swamp (Ludlow Swamp), approximately 7 km west of the project area and the Vasse-Wonnerup Wetlands approximately 11 km to the west.

2.3 Riverine Condition Assessment The relative abundance (proportional surface area of dominant habitat types) and condition of dominant aquatic habitats at each site was recorded. A photographic record of each site was established for any ongoing monitoring. This will help detect changes in site condition, such as sedimentation or erosion, riparian condition etc. Location of each site was confirmed by GPS. The Foreshore Assessment techniques of Pen & Scott (1995) and WRC (1999) were used for the study area and applied at each riverine sampling location. These techniques are designed for rapid, qualitative assessment of bank condition (e.g. erosion characteristics), vegetation status (e.g. health & percentage cover) and presence of weed species. Based on the assessment, an overall Environmental Rating is then determined for each site. The current study did not include a detailed assessment of riverine flora. Comprehensive flora and vegetation surveys have been undertaken independently.

2.4 Physico-chemistry Measurements of physico-chemical parameters were made in conjunction with the aquatic fauna sampling. Parameters recorded are summarized in Table 2. Measurements of temperature, dissolved oxygen, conductivity and pH were made in situ between the hours of 0900 and 1600 using portable WTW field meters. A Secchi disk was used for field determination of water clarity. Undisturbed water samples were collected from sites 1, 3, 4 and 6 for laboratory analyses of ionic composition, total suspended solids, metals, total petroleum hydrocarbons and nutrients. Samples collected for nutrients and metals were filtered through 0.45 µm Millipore nitrocellulose filters and samples for metals were acidified in the field. Nitrates and soluble reactive phosphorus were recorded as an indication of the bio-available component of phosphorus in the water column.

Table 2. Physico-chemical parameters measured at each site.

Parameter Code Units

Water depth (maximum & average) Max. & Ave. depth m Wetted width (average) Ave. width m Channel width (average) Ave. channel width m Dissolved oxygen DO % % Dissolved oxygen DO mg/L mg/L Electrical Conductivity ECond
S/cm pH pH pH units Water Temperature Temp oC Water clarity Secchi cm

Water Hardness (CaCO3) Hardness mg/L Aluminium Al mg/L Alkalinity Alkalinity mg/L Arsenic As mg/L Boron B mg/L Barium Ba mg/L Calcium Ca mg/L

Carbonate CO3 Mg/L Cadmium Cd mg/L Chloride Cl mg/L Cobalt Co mg/L Chromium (unfiltered – as total) Cr mg/L

2 International treaty The Convention on Wetlands, signed in Ramsar, Iran, in 1971, on the conservation and wise use of wetlands and their resources, including migratory birds. Currently there are 1526 wetland sites nationally and internationally, covering 129.5 million hectares, listed under the Ramsar List of Wetlands of International Importance. http://www.ramsar.org/

______Wetland Research & Management ______5 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Parameter Code Units

Copper Cu mg/L Iron Fe mg/L

Bicarbonate HCO3 mg/L Potassium K mg/L Manganese Mn mg/L Mercury Hg mg/L Molybdenum Mo mg/L

Nitrate NO3-N mg/L

Ammonia NH3-N mg/L Nitrogen – total Kjeldahl TN mg/L Sodium Na mg/L Nickel Ni mg/L Lead Pb mg/L Soluble reactive phosphorus P-SR mg/L Phosphorus - total TP mg/L Selenium Se mg/L

Sulphate SO4 mg/L Vanadium V mg/L Zinc Zn mg/L Total dissolved solids – 180 deg. C TDS mg/L Total suspended solids TSS mg/L Total petroleum hydrocarbons TPH µg/L

All samples were stored on ice in the field and frozen as soon as possible for subsequent transport to the laboratory. All laboratory analyses were conducted by the Natural Resources Chemistry Laboratory, Chemistry Centre, WA (a NATA accredited laboratory).

Water depth was measured using a graduated pole while channel width and wetted width was measured with a 50m tape measure. Dominant habitat substrates were visually appraised (by surface area) for mineral or other (e.g. vegetation, organic detritus) material. Extent of bank erosion and channel down- cutting was qualitatively assessed. Descriptions of overall stream condition were based on categories outlined in WRC (1999).

2.5 Macroinvertebrate Fauna The macroinvertebrate fauna (i.e. fauna retained by a 250 µm aperture mesh) was targeted as it typically constitutes the largest and most conspicuous component of aquatic invertebrate fauna in lotic (flowing) waters. At each site, sampling aimed to maximise the number of taxa collected, by sampling as many in-stream habitats as possible over a total 50 m within a (maximum) 200 m reach of river. This included sampling reed/rush beds, draped vegetation, woody debris, open water column and benthic sediments (cobbles, gravels, sand & silt). Sampling was conducted with a 250µm mesh net to selectively collect the macroinvertebrate fauna. This protocol is the same as that used by WRM for long-term monitoring of the Buntine Marchagee and Muir-Unicup Recovery Catchment wetlands on behalf of the Department of Conservation and Land Management (CALM).

Samples were preserved in 70% ethanol and returned to the laboratory for sorting under low power microscope to remove animals. All taxa were identified to the lowest level (species, where possible) and enumerated to log10 scale abundance classes (i.e. 1 = 1 - 10 individuals, 2 = 11 - 100 individuals, 3 = 101-1000 individuals, 4 = >1000). In-house expertise was used to identify invertebrate taxa using available published keys and through reference to the established voucher collections held by the Aquatic Research Laboratory, School of Animal Biology, The University of Western Australia. External specialist taxonomic expertise was sub-contracted on an ‘as needs be’ basis to assist with specific groups. This included Dr Don Edward (UWA) for Chironomidae (non-biting midges) and Dr

______Wetland Research & Management ______6 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Mark Harvey (Western Australia Museum) for Acarina (aquatic mites). The existence of rare, restricted or endemic species was determined by cross-referencing taxa lists for each site with the UWA database, the CALM Wildlife Conservation (Specially Protected Fauna) Notice 2005 and with the IUCN Red List of Threatened Species (2004).

2.6 Fish and Crayfish A range of sampling techniques (sweep sampling, baited traps, electrofisher & visual survey) was used to collect as comprehensive a list of fish and freshwater crayfish as possible. Box traps were set overnight in deeper pools and cleared each morning. At each site the channel was selectively fished using a Smith-Root Model 12-B battery powered backpack electrofisher. Electrofishing was typically performed in an upstream direction, shocking in all habitats with the intention of recovering as many species as possible. Fish and larger crayfish were identified in the field and all native species released alive. Fish nomenclature followed that of Allen et al. (2002).

2.7 Frogs and Waterbirds Qualitative sampling for tadpoles was conducted by random sweeps of a 1 mm mesh pond-net and baited mesh box traps used to sample fish. A small number (<10) of individual tadpoles was retained and preserved in 4% buffered formalin for later identification. Opportunistic surveys of adult frogs were conducted in conjunction with tadpole surveys, by comparing any calls heard on the day of sampling with audio files for south-west species.

The total number of each species of water-dependent bird (i.e. waterfowl, water hen, herons, egrets, cormorants etc) was recorded from the vicinity of each site when sampled for fish and invertebrates. Any evidence of recent usage by waterbirds was also recorded. There is a tendency to ‘drive’ birds along the watercourses when sampling, so every attempt was made to avoid counting birds twice and not to count birds moving from one site into the next. Species present were detailed against conservation significance and listings such as JAMBA and CAMBA3.

2.8 Data Analysis All data, including taxonomic lists, abundance data, physico-chemical measurements and GPS locations were entered onto a Microsoft ‘Excel’ spreadsheet and a copy lodged with the contracting organisation for future reference. The occurrence of each species as collected by all sampling methods was tabulated.

2.8.1 Univariate Analysis Using sites as replicates, one-way analysis of variance (one-way ANOVA) was used to test for significant differences in community parameters (i.e. species richness) and physico-chemical parameters between control and exposed sites.

2.8.2 M ultivariate Analysis Multivariate pattern analysis was performed using ordination techniques to group sites according to total invertebrate faunal assemblages on both presence/absence and abundance data sets. This approach indicates groups of similar/dissimilar sites based on fauna. Data were analyzed using multivariate procedures from the PRIMER (v5) software package (Clarke & Gorley 2001). Up to four levels of analysis were applied to the data: 1. Describing pattern amongst the fauna assemblage data using cluster and ordination techniques based on Bray-Curtis similarity matrices. The clustering technique uses a hierarchical agglomerative method where samples of similar assemblages are grouped and the groups

3 Bilateral migratory bird agreements with Japan (JAMBA) and China (CAMBA).

______Wetland Research & Management ______7 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

themselves form clusters at lower levels of similarity. A group average linkage was used to derive the resultant dendrogram. The ordination method used was Multi-Dimensionsal Scaling (MDS) (Clarke & Warwick 2001). Ordinations were depicted as two-dimensional plots based on the site by site similarity matrices. 2. For any groups found in 1 or selected a priori (i.e. exposed versus control), Analysis of Similarity (ANOSIM) – effectively an analogue of the univariate ANOVA – was conducted to determine if groups were significantly different from one another. The ANOSIM test statistic reflects the observed differences between groups (e.g. control vs exposed) with the differences amongst replicates within the groups. The test is based upon rank similarities between samples in the underlying Bray- Curtis similarity matrix. The analysis presents the significance of the overall test (Significance level of sample statistic), and significance of each pairwise comparison (Significance level %), with degree of separation between groups (R-statistic), where R-statistic >0.75 = groups well separated, R-statistic >0.5 = groups overlapping but clearly different, and R-statistic <0.25 = groups barely separable. A significance level <5% = significant effect/difference. 3. The SIMPER routine was used to examine which taxa were contributing to the differences of any groups that were found to be different according to the ANOSIM procedure or otherwise found to be separated in cluster or ordination analyses. 4. The relationship between the environmental and biotic data was assessed in two ways: - For visualization, the numeric value of key environmental data were superimposed onto MDS ordinations, as circles of differing sizes – so-called ‘bubble plots’. - The BIOENV routine was used to calculate the smallest subset of environmental variables explaining the greatest percentage of variation in the ordination patterns based on fauna.

Environmental data were analysed using Principal Components Analysis (PCA) to discern patterns, gradients and similarities in water quality amongst the sampling sites. PCA transforms a number of (possibly) correlated variables into a (smaller) number of uncorrelated variables called principal components. The first principal component accounts for as much of the variability in the data as possible and each succeeding component accounts for as much of the remaining variability as possible. The starting point for a PCA is a correlation matrix based upon Euclidean distance. PCA ordinations were performed using untransformed data. For metal concentrations, below the detection limit data were assigned values of half the detection limit for all analyses. For visualization, the numeric value of key environmental variables was superimposed onto the PCA ordinations, as ‘bubble plots’.

Note: The entire suite of water quality variables was only measured for four sites (sites 1, 3, 4 & 6) representative of four different reaches/habitat conditions within the catchment. Therefore, for PCA where a full suite of variables was required for all sites, measured water quality was taken as indicative of water quality for all sites within each reach: reach 1 (Site 1), reach 2 (sites 2 & 3), reach 3 (sites 4 & 5) and reach 4 (sites 6, 7 & 8).

______Wetland Research & Management ______8 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

3. RESULTS AND DISCUSSION 3.1 Riparian and In-stream Habitat Condition Riverine condition of all sites was considered degraded due to historic pastoral practices and unrestricted livestock access to the natural waterbodies. Plates 1 - 9 and the cover photo illustrate the conditions at sites at the time of sampling. The riverine landscape was typical of rural regions throughout the south-west. Channels were characterised by poor bank stability with extensive erosion – bank slumping, channel widening, bed down-cutting and (often) extensive sedimentation (Table 3). There was active head-cutting in Tiger Gully at the confluence with the old Ludlow River (Site 5).

In-stream vegetation was typically dominated by filamentous algae (often covering up to 30% of bed substrates) with only isolated clumps of submerged native macrophytes and emergent sedges. Terrestrial native understorey vegetation was, at best, sparse; the understorey dominated by pasture grasses (e.g. Kikuyu) and weeds, which for the most part was providing a degree of protection against excessive bank erosion. Along the river, overstorey vegetation ranged from open to moderately dense mixed woodlands of marri (Corymbia calophylla), blackbutt (Eucalyptus patens) and peppermints (Agonis flexuosa) with some river red gums (E. rudis) and paperbarks (Melaleuca rhaphiophylla). With the exception of the Ludlow conservation area and to a lesser degree, the Capel River, most trees observed were mature, with little evidence of recruitment. An environmental rating of ‘moderate’ was assigned to the Ludlow conservation area and Capel River site, but all other sites were rated as ‘poor’ to ‘very poor’ on the basis of health and percentage cover of remnant vegetation and the degree of bank/bed erosion (Table 3). The total width of the forested riparian zone in the conservation area varied from around 50m to 100m. If allowed to regenerate over this area, native vegetation should provide a good buffer between the river and surrounding agricultural properties. Similarly, recent fencing by landowners along the Capel River will assist the rehabilitation of native vegetation in a strip some 20m wide either side of the river. The regional ecological value of the remnant riparian vegetation and in particular the conservation wetland area was considered to be high, given the extensive (historic) clearing of native vegetation for agriculture. The remnant vegetation is likely to provide at least some of the energy (‘food source’) that drives many aquatic processes (e.g. food webs) as well as providing food, shade and shelter for both terrestrial and aquatic fauna. Where ever possible, the trees and few remaining perennial shrubs should be conserved to support food webs and to help stabilise the river banks against further erosion.

Plate 1. Site 1 (lot 3739), view west toward confluence Plate 2. Site 3, Ludlow River on lot 2, uppermost with Tiger Gully (entering from the right & flowing out control site. Paddock with mature overstorey trees to left), downstream from Yoganup discharge point and remnant native shrubs lining stream banks. (YOD) and upstream of Yoganup West mine. Sand bar formation evident on inside meander. Tannin stained waters of Tiger Gully can be seen flowing round toe of sand bar.

______Wetland Research & Management ______9 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Plate 3. Site 2 (YWLRU), Ludlow River on lot 1392. Plate 4. Site 4 (TGD; lot 1188), Iluka’s Tiger Gully Immediately upstream from Ludlow diversion point rehabilitation area downstream from Yoganup West. above Yoganup West mine. Meanders have been constructed in the channel to reduce flow and create pool-run sequences. Banks have been replanted with native seedlings.

Plate 5. Site 5, Tiger Gully at confluence with old Plate 6. Site 6, Ludlow River, lot 2015; conservation Ludlow River channel (lot 1392). Paddock with sparse area immediately east of Warns Road. Lot of eucalypt native riparian vegetation. Banks downcut. recruits along banks, but understorey dominated by weeds.

Plate 7. Site 8, Ludlow River conservation area Plate 8. Site 7, Ludlow River conservation area. Most bordering lots 3229 & 2012. Bank slumping and downstream site on lots 3229/2012, close to Capel- channel widening particularly evident in foreground. Tutunup Road. Bank slumping and extensive tree root scour evident. Understorey weed-dominated.

______Wetland Research & Management ______10 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Plate 9. Site 9 on the Capel River (lot 3299), January 2006; (above) view from farm bridge looking up- riverwith understorey vegetation dominated by pastures grasses, e.g. kikuyu; (right) view down-river showing abundant in-stream woody debris and fringing peppermints with riffle area in foreground flowing into pool formed by snags across channel.

Table 3. Summary of riverine condition. Foreshore condition and environmental ratings based on those of WRC (1999).

Site Description Foreshore Condition Environmental Rating 1 Tiger Gully, near confluence d/s of Iluka’s discharge point (YDP). Stream flows thro’ C1: erosion prone; Very poor broad, flat seasonal wetland area with only scattered remnant paperbarks over pasture understorey weeds only grasses and native sedges/rushes. Lemna sp. lines channel. Waters highly tannin stained. Bank height very low but very steep >60°; bed substrates sand & rock with clay banks; tho’ soil cohesion appears good along banks the channel is heavily sedimented at confluence (sand to 30 cm deep). Fenced – no cattle access. 2 Upper Ludlow River immediately u/s of YWLRU sampling point. Open to moderately C1: erosion prone; Poor dense eucalypt overstorey. Understorey all pasture & weeds. Banks steep to very understorey weeds only steep 45-60+°; soil clayey loam; channel sedimented to 20cm deep & downcut by 1.5 – 2m; series of runs & pool habitats but both sedimented; 20% submerged macrophyte cover but only sparse large woody debris (LWD) & leaf packs. Lot of small snags. Bed substrates of fine silt over sandy loam & clay. Fenced – limited cattle access. 3 Upper Ludlow River, opposite main house. Open to mod-dense eucalypt overstorey C1: erosion prone; Poor and tall shrub layer (Myrtacea spp. & Acacia sp.) providing 50% shade cover; understorey weeds only understorey otherwise all pasture; Banks steep 45-60°; poor soil cohesion (sandy clays) in this section; river channel is series of riffles & pools; sandy loam & clay bed substrates; <10% sub-merged macrophyte cover; 20-30% leaf pack cover; some LWD. Channel erosion similar in extent to that at Site 2. Fenced – only limited cattle access. 4 Iluka’s rehabilitation site on Tiger Gully; streamlined – meanders; no shade cover as C2-C3: eroding Poor – Very poor vegetation only recently replanted; banks moderately steep 10-45°; poor soil cohesion (but streamlining (but streamlining (sandy) but erosion-control mats line bed & banks; lot of filamentous algae (25% of in- commenced) commenced) stream habitat); <5% macrophytes; sand dominated with broad, deep sand bars on meander bends. Fenced – no cattle access. 5 Confluence of old Ludlow River and Tiger Gully (d/s of Site 4). Open overstorey of C3: eroding Very poor eucalypts & paperbarks over pasture. Reach has been straightened to act as ag. drain.

______Wetland Research & Management ______11 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Another drain enters from south-west. Cattle access. Banks low but very steep >60; general soil cohesion good. Active head-cutting has formed small waterfall with 1 m deep scour pool at base; otherwise channel very shallow; clay soils dominate overlain with gravel runs; 70% filamentous algal cover & blue-green mats; anoxic; 20-30% macrophyte cover (mainly Lemna sp.) 6 Conservation category wetland area in mid-reaches of Ludlow River. River here B3-C1: good over- Moderate receives lot of input from seasonally inundated paddock to the NE (P. Nelson pers. storey but channel cuti.). Pasture species encroaching into channel; 30% cover of filamentous algae in- generally erosion prone stream; channel downcut by 1.5m on some outer meander bends. Mod-dense eucalypt with localised areas of overstorey (regrowth) and tall shrub layer (Myrtacea spp.) along both banks providing undercutting & bank 50% shade; lot of tree recruits; understorey otherwise all pasture; arum lily. Banks slumping steep 45-60°; poor to moderate soil cohesion; river channel is series of gravel runs & pools; gravel-sand bed substrates; <10% macrophyte cover; 20-30% leaf pack cover; little LWD – appears to be de-snagged; Fenced – only limited cattle access. 7 Conservation category wetland area in mid-reaches of Ludlow River, similar to Site 6. B3-C1: good over- Moderate Mod-dense to dense overstorey of young eucalypts (lot of recruitment) and storey but channel peppermints along both banks. Understorey dominated by pasture species with only generally erosion prone scattered native shrubs & sedges. Banks moderately steep 10-45°; soil cohesion would with localised areas of appear good (clayey loam & gravel) but channel actively eroding – bank slumping & undercutting & bank bed down-cutting (to 2m on outer meanders); deeply scoured tree roots along banks; slumping lot of exposed (~20%) soil along banks. Fenced – limited cattle access. 8 Conservation category wetland area in mid-reaches of Ludlow River, similar to Sites 6 B3-C1: good over- Moderate & 7. Mod-dense to dense overstorey of young eucalypts (lot of recruitment) and storey but channel peppermints along both banks. Understorey dominated by pasture species with only generally erosion prone scattered native shrubs & sedges. Some Vallisneria sp and a lot of ?Rupia sp. (20% with localised areas of cover). Banks very steep >60°; channel actively eroding – bank slumping & bed down- undercutting & bank cutting (to 2m on outer meanders); lot of exposed (~20%) soil along banks but good slumping general soil cohesion; sand bars/deposition (30 cm deep) on meanders. Fenced – limited cattle access. 9 Capel River along northern boundary of Cloverdale Project Area. Open to moderately C1: good overstorey Moderate dense marri & peppermint overstorey. Understorey predominantly pasture & weeds but channel generally with only scattered tall shrubs. Banks very steep >60°; good soil cohesion (clayey erosion prone with loams) but channel sedimented to 20cm deep; bank slumping & downcutting by 1.5 – localised areas of 2m; abundant large woody debris (LWD) & 20-30% in-stream leaf packs. Good undercutting & bank heterogeneity of in-stream habitat – deep pools, gravel runs, sand bars, snags, trailing slumping vege. Lot of small snags. ~5% in-stream cover of Vallisneria & ~5% emergent rush/sedges; Bed substrates of fine silt over sandy loam & clay. Recently fenced – limited cattle access only at stock watering point.

3.2 Physico-chemistry Weather conditions over most of the sampling period consisted of moderate cloud cover, occasional light showers during the day, relatively warm air temperatures and light winds. Relatively high rainfall was experienced during January 2006 sampling in the Capel River. Water levels in the Ludlow system were higher than usual for the time of year owing to atypically high spring rainfall. However, water levels in the Capel River were considered (Tom Hutton, pers. comm.) typical of summer baseflow conditions. The physico-chemical characteristics of each site are detailed in Appendix 1 and ANZECC/ARMCANZ (2000a) guidelines for the protection of aquatic ecosystems and for livestock drinking water are given in Appendix 2. In most instances, current guideline values for aquatic ecosystems are lower than those recommended for livestock drinking water (refer Appendix 1). Though sites lay within (largely) cleared agricultural lands, values for most of the physico-chemical variables measured in situ (Table 4) were still within the ranges expected for only ‘slightly to moderately disturbed’ aquatic ecosystems (ANZECC/ARMCANZ 2000a). None of the parameters measured exceeded recommended maxima for livestock drinking water.

______Wetland Research & Management ______12 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

In the Ludlow system, salinity (& associated ion concentrations), water hardness and alkalinity tended to be greatest near the Yoganup discharge point (YDP, Site 1, Figure 2). Also evident was a slight gradient in salinity, with values tending to increase longitudinally down the catchment (Figure 2). While all waters were fresh (411 - 819 JS/cm), levels exceeded the maximum trigger value of 300 JS/cm (ANZECC/ ARMCANZ 2000a) for aquatic ecosystems in south-west rivers. Salinity in the Capel River in January 2006 was 1,115 JS/cm. The composition of major ions in the Ludlow Econd (uS/cm) Alkalinity CaCO3 (mg/L) system was typically dominated by sodium and 1000 200 + 2+ + 2+ - 2- chloride (Na >Mg >K > Ca : Cl>SO4 900 180 - >HCO3 ). However, the concentration of 800 160 s

s calcium and sulphate ions was much greater at 700 140 e n . d d 600 120 r YDP (Site 1); 12% and 33% of the total major n a o H

C 500 100 & cation and anion concentrations, respectively. E

y t 400 80 i n

i The ionic composition of waters will be l a

300 60 k l influenced by rain-borne salts (i.e. wind blown A 200 40 dusts) and geology (e.g. weathering of soils) of 100 20 the catchment. However, the composition 0 0 3 2 1 4 5 6 8 7 over the warmer months, particularly in Site shallow reaches, will be altered by evapo- concentration and precipitation of less soluble Figure 2. Changes in salinity (as Econd.) within the salts, such as calcium carbonate and Ludlow /Tiger Gully catchment. magnesium sulphate (Hart & McKelvie 1986). The ionic composition of inland waters in Australia is known to vary widely. Only waters at the YDP (Site 1) were considered hard, i.e. containing >60 mg/L CaCO3. There are currently no aquatic ecosystems guidelines for major cation/anion concentration and all recorded concentrations were well below guidelines for livestock drinking water.

Waters at most sites appeared visually turbid and slightly tannin stained, but Secchi depth was usually greater than maximum water depth and TSS values were relatively low <6 mg/L. Maximum TSS concentrations were recorded from below YDP (Site 1) and Site 6, compared to <2 mg/L TSS at sites 3 and 4 (TGD)). Water pH varied from slightly acidic to slightly alkaline (6.14 - 7.36).

The turbidity, salinity, alkalinity and hardness of water in the receiving environments of sites 4 to 8 will depend on water quality in both the upper Ludlow River (e.g. Sites 2 & 3) and Tiger Gully (e.g. Site 1). Higher levels at sites 4 to 8, relative to sites 2 and 3 may reflect shandying of poorer quality water entering from Tiger Gully / YDP. Plots of water quality data collected by Iluka at YDP over the period 1999-2005 show what may be a slight upward trend in salinity (Econd.) and sulphate levels post November 2003 and a downward trend in pH during 2005 (Figure 3). The changes are minor and within the limits imposed by the current Yoganup/Yoganup West licence (ECond. 1,250 µS/cm & pH 5.5 - 9.0 pH). Though there are no comprehensive records for Tiger Gully prior to mining, levels would still appear to be within the seasonal ‘background’ range known for the Ludlow River and nearby Capel River (see Section 3.2.2 below). However, levels should continue to be monitored to determine if they do indeed represent actual trends and not just short-term fluctuations.

Water temperatures at all sites were within the range 17.6 to 22.8 °C reflecting differences in water depth (i.e. shallower waterbodies have greater daily ranges in temperature), time of day when sampled (sites sampled in the morning will likely be cooler than those sampled in the afternoon) and time of year. Dissolved oxygen levels were unexpectedly high at many sites ranging from 75% to 104% (super- saturated), possibly reflecting in-stream algal productivity.

______Wetland Research & Management ______13 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

YDP Econd (uS/cm) 1400

1200

1000

800

600

400

200

0 9 3 2 2 4 4 5 5 4 0 0 0 1 1 1 2 3 3 4 5 3 4 5 0 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 / / / / / / / / / / / / / / / / / / / / / / / 6 1 8 6 6 1 8 8 1 9 7 9 7 0 7 6 5 4 0 9 5 9 7 / / / / / / / / / / / / / / / / / / / 1 1 1 1 2 9 / 1 1 4 3 6 4 2 5 5 1 / 5 8 / 5 / 1 5 6 4 2 2 7 1 2 2 1 2 1 2 1 2 1 2 1 8 3 3 1 1 1 1

YDP Sulphate (mg/L) 160

140

120

100

80

60

40

20

0 1 0 2 3 3 1 2 4 3 4 4 4 5 5 1 2 9 3 0 0 4 5 5 0 0 0 0 0 0 0 0 0 9 0 0 0 0 0 0 0 0 0 0 0 0 0 / / / / / / / / / / / / / / / / / / / / / / / 1 9 0 9 0 1 9 6 7 8 9 1 5 6 7 4 5 7 6 7 6 8 8 / / / / / / / / / / / / / / / / / / / 1 1 1 1 / / 4 1 / 1 8 2 5 4 6 / 1 4 5 9 3 1 2 5 6 5 5 7 3 8 3 2 2 1 2 2 2 1 1 2 1 2 1 1 2 1 1 1 1

YDP pH 8.5

8

7.5

7

6.5

6

5.5

5 9 2 1 3 0 1 3 5 3 5 4 4 0 2 0 2 4 1 3 4 5 5 4 0 0 0 0 0 0 0 9 0 0 0 0 0 0 0 0 0 0 0 0 0 0 0 / / / / / / / / / / / / / / / / / / / / / / / 9 6 7 6 6 8 9 1 8 1 0 5 6 5 1 9 9 4 7 7 0 8 7 / / / / / / / / / / / / / / / / / / / 1 1 1 1 9 3 8 / / 5 / 5 1 1 1 4 / 6 2 4 5 1 2 4 5 5 6 3 3 8 2 2 2 1 1 1 2 1 2 7 2 1 1 2 2 1 1 1 1 Figure 3. Iluka water quality data showing longer-term changes in salinity (Econd. µS/cm), sulphate (mg/L) and pH at Site 1 (YDP).

______Wetland Research & Management ______14 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

In the Ludlow system, nutrient concentrations ranged from 0.2 - 0.4 mg TN/L, 0.01 - 0.19 mg NO3/L and 0.02 - 0.03 mg TP/L. Soluble reactive phosphorus (SRP) was below the detection level of 0.01 mg/L for filterable reactive phosphorus (FRP). While NO3 concentrations at control Site 3 and downstream Site 6 approached ANZECC/ARMCANZ (2000a) ecosystem guideline values, other nutrient readings were unexpectedly low given the past land management practices and unrestricted access of livestock to waterways. The relatively high spring rainfall likely accounted for the lower than expected nutrient levels. It should also be noted that spot measurements of nutrients (or of metals, hydrocarbons, salts and suspended solids) are not necessarily indicative of total nutrient loads.

All hydrocarbon and most metal concentrations (As, B, Co, Cr, Cu, Hg, Mo, Pb, Se & V) were below detection limits. All metal concentrations were below livestock drinking water guidelines (ANZECC/ ARMCANZ 2000a), however Al and Cd exceeded aquatic ecosystem guideline limits at some sites and Zn exceeded guidelines at all sites. It must also be noted that detection limits for Cr and Cu were between two to five times greater than aquatic ecosystem guidelines. While Al and Cd were only slightly above ecosystem guideline limits at control Site 3, they were much greater than levels recorded at all other sites; up to ten times greater for Al. Zinc levels exceeded aquatic ecosystem guidelines at all sites, but in particular at Site 1 (near YDP). Even when water hardness was taken into account, zinc levels at YDP were more than twice the recommended maxima. The source of the elevated metals at Site 3 was not considered due to mining as this site lay upstream from both Yoganup West and Yoganup mine sites. The high level of zinc at all sites was taken as indicative of the geology of the area, though mining may have resulted in the relatively greater concentrations at Site 1. There were no baseline data available for zinc concentrations.

Even though elevated, it is unknown what proportion of the measured dissolved metals was bio- available or unavailable through complexing with dissolved organic carbon (e.g. tannin). ANZECC/ARMCANZ (2000a) recommends use of techniques such as DGTs (Diffuse Gradients in Thin Films) as a speciation measurement to provide a better estimate of the bio-available metal concentration if the dissolved metal concentrations exceed the guideline trigger values.

Diffuse Gradients in Thin Films (DGTs) The DGT technique was first developed in 1994 as a time averaged, in situ speciation measurement of heavy metals in waters. Since its introduction it has been validated in the field for the determination of metals in fresh and seawater, and more recently in estuarine waters. The DGT technique is based on a simple device, which accumulates metal ions in a well-defined manner from solution. Soluble species diffuse through a diffusive layer of known thickness in which a concentration gradient is maintained. Behind the diffusive layer is a binding layer in which reactive metal species are bound. The mass of accumulated metal is measured following retrieval and is used to calculate the average concentration of DGT labile metal species in the bulk solution over the deployment time. As the device does not accumulate the major ions that cause interference with the measurement, the measurement does not suffer the degree of interference associated with the direct analysis of waters.

3.2.1 Patterns in Physico-chemical Data The Capel River was not included in statistical analyses of baseline data, as it was surveyed at a different time of year and at only one reach. It has been included in plots provided in Appendix 3c merely to illustrate differences between rivers / sampling occasions.

Though physico-chemistry varied between individual sites within the Ludlow system, there were no significant differences (ANOVA; df = 1, F = 0.4028, p = 0.8411) between upstream Cloverdale controls (sites 1, 2, 3, & 4) and downstream exposed sites (sites 5, 6, 7 & 8) for most of the measured environmental variables. Nor was there any difference between controls outside the influence of existing Yoganup and Yoganup West mines (i.e. sites 2 & 3) and control sites downstream from

______Wetland Research & Management ______15 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Yoganup West (sites 1 & 4) (ANOVA; df = 1, F = 0.1020, p = 0.7498). Therefore Ludlow River and Tiger Gully sites selected as controls for Cloverdale should generally serve as good pre-mine reference sites against which to compare any future changes in water quality of the receiving environment. Salinity, however was significantly higher at exposed sites (5, 6, 7 & 8) when compared against control sites 2, 3 and 4 (ANOVA; df = 1, F = 9.898, p = 0.0255), reflecting the longitudinal gradient in Econd. and TDS mentioned above.

Principal Components Analysis (PCA) on the eight samples reduced the entire physico-chemical dataset to five principal components, with 92.2% of the total variation being explained by PC1, PC2 and PC3 (Table 4 & Appendix 3a-b).

Though ANOVAs showed no significant differences in overall water quality, PCA ordination did indicate patterns between sites. Physico-chemical parameters showed a distinct separation of Site 1 (near YOD) from the other seven sites (Figure 4a) sampled in the Ludlow/Tiger Gully system. Control sites 2 and 3 also appeared to differ from other sites (4, 5, 6, 7 & 8; Figure 4a), all of which were exposed to existing Yoganup and Yoganup West developments. PCA analysis was rerun using sites 2 - 8 only (with Site 1 considered an outlier) (Figure 4b) and resulting Eigen values were similar to those for the total data set (Table 4). In both ordinations, uppermost control sites 2 and 3 on the Ludlow River were similar to each other (Figures 4a & b), but relatively dissimilar to downstream control site 4 (Iluka’s rehabilitation site, TGD). Site 4 was most similar to Site 5, which lay within cleared paddocks immediately east of Warns Road at the confluence of Tiger Gully and the old Ludlow River. These sites however, were clearly different from those in the downstream forested Ludlow River conservation area (sites 6 - 8) (Figure 4b)

For the total data set (Figure 4a), variables contributing to PC1 were salinity (Econd. & TDS), Ca, Cl, K and Mg ions, water hardness and nitrates. Site 1 separated from all other sites along this principal component due to its higher ion levels (Ca, Na, K, Mg & SO4), hard water and very low nitrate levels. Separation of other sites primarily reflected the longitudinal salinity gradient within the catchment. Variables contributing to PC2 (Figure 4a) were Ni, Fe, Mn, Al, Cd, pH and total nitrogen. Minimum Fe levels were recorded at Site 1 and maximum levels at Site 6, while maximum Ni levels were recorded at Site 1. Higher concentrations of Zn at Site 1 also separated this site from all others along PC3 (not shown).

For the ordination without Site 1 (Figure 4b), variables contributing to separation of sites along PC1 were relatively high levels of Al, Cd and Ni at Site 3 and relatively low ion concentration and water hardness. Variables contributing to PC2 (Figure 4b) were low levels of Zn and ammonia at Site 4 and slightly higher levels of TP at Site 6.

The results may reflect some impact of mining at Yoganup/Yoganup West on downstream water quality, as sites 2 and 3 which were outside the influence of mining were clearly distinct from downstream sites and from Site 1 (YOD). If the relatively higher Ni and Zn levels at YDP are associated with mining, then effects would appear to be highly localised as levels of these metals are much lower at Site 4, even though waters at sites 2 and 3 were metal-rich by comparison. However, as mentioned above, mining cannot be discounted as contributing to the increased salinity (& associated ions) in the receiving environments of sites 4 to 8, though increases are still within licence limits. Other possible influencing factors which should also be considered are the unique, broad, flat wetland and underlying geology at Site 1 resulting in increased salts, water hardness, Ni and Zn levels in the shallow waters of the drainage line.

Similar (though less distinct) separation of sites was also evident for PCA analyses conducted on only those physico-chemical data collected in situ (i.e. DO, temp., Econd, TDS, pH, depth & width) with Site 1 separating from all others on the basis of higher salinity and sites 2 and 3 differing from each other

______Wetland Research & Management ______16 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06 due to higher recorded DO levels at Site 2 and slight variations in pH and water temperature (refer Appendix 3d).

Table 4. Eigen values, percent of variance and cumulative percent of variance explained by the five components from PCA performed on standardised physico-chemical parameters for the total data set and for sites 2 – 8 only.

Total phys-chem data set – all sites Phys-chem data set for sites 2 – 8 only PC Eigen % Variation Cumulative % PC Eigen % Variation Cumulative Values Variation Values % Variation 1 16.12 52.0 52.0 1 18.92 61.0 61.0 2 7.11 22.9 74.9 2 8.40 27.1 88.1 3 5.35 17.3 92.2 3 2.68 8.6 96.8 4 1.80 5.8 98.0 4 0.52 1.7 98.4 5 0.39 1.3 99.2 5 0.33 1.1 99.5

(a) Total phys-chem data for Ludlow system

4 1 3 2 3 2

2 C

P 1

0 5 -1 4

-2 67 -3 8 -8 -6 -4 -2 0 2 4 6 8 10

PC1

(b) Phys-chem data for sites 2 – 8 only

4 4 5 3

2

2

C 2

P 1

0 3

-1

-2 8 -3 6 7 -8 -6 -4 -2 0 2 4 6 8 10

PC1 Figure 4. Two-dimensional PCA ordination plots of physico-chemical variables for (a) all sites surveyed in November 2005, and (b) sites 2 – 8 only. Code: p control sites; r controls for Cloverdale but exposed to existing Yoganup & Yoganup West; p exposed sites for Cloverdale, Yoganup West and Yoganup.

______Wetland Research & Management ______17 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

3.2.2 Comparisons with Known Water Quality Data for Other Systems in the Area There is little specific published information about the quality of surface waters in the study area, though there are a number of unpublished reports for the Capel River (refer White & Comer 1999). Some nutrient data for the Ludlow River is also available on the GeoCatch website: http://www.geocatch.asn.au/pages/framesaction.html?=action.html#RAPs.

Poor water quality in the Capel River is an on-going problem, which local LCDC groups are trying to address through river foreshore restoration, dairy effluent management and restricting stock access. These problems are also an issue for the Ludlow River, with previously recorded seasonal maxima TP levels of around 0.2 mg/L and TN of 2.5 – 3.0 mg/L. Levels recorded during the current study were slightly higher than this at some sites; i.e. 0.03 mg TP/L at Site 6 and 0.4 mg TN/L at Site 4 (TGD). In comparison with these data, water quality parameters measured in the Capel River have a typical range of: Econd. 300 - 1,400 µS/cm, pH 5.5 - 8.5, winter-time turbidity 180 NTU, TP <0.1 mg/L (but occasionally up to 0.4 mg/L), NH3 <0.5mg/L and TN <0.45 mg/L. All parameters show strong seasonality. Algal blooms have been recorded in the Capel River on several occasions as well as temperature inversions over summer which resulted in anoxic bottom muds rising to the surface (White & Comer 1999). Salinity levels show strong seasonality with the river moderately brackish over much of the year, but increasingly saline following the first winter flush then becoming fresh toward the end of winter. Though nutrients and turbidity in the Capel River were not measured during the current study, salinity and pH in January 2006 were within these previously recorded ranges. In the Ludlow system, the relatively low conductivity and circum-neutral pH recorded during November 2005, was likely due to the direct influence of rainwater and a wetter than usual spring. Like the Capel River, salinity in the Ludlow system likely increases as the dry season progresses, coupled with temporarily higher values following the first of the seasonal rains. Seasonal effects resulting from reduced flows in the Ludlow River are illustrated in Table 5 below. In January 2006, control sites 2 (YWLRU) and 3 on the Ludlow River were re-surveyed as part of aquatic ecosystems surveys for the Yoganup 215 project (see WRM 2006a). At this time, there was no flow and the river was reduced to a number of permanent and semi-permanent pools. Major changes in water quality were an increase in salinity levels (particularly at Site 2) and decreased dissolved oxygen concentrations (Table 5).

Table 5. Seasonal variations in water quality recorded at control sites on the Ludlow River between a period of flow (spring 2005) and no flow (summer 2006).

Site Date Temp DO% (mg/L) pH ECond. (S/cm) TDS (mg/L) 2 (YWLRU) Nov. 2005 19.8 118 (12.6) 6.91 461 237 2 (YWLRU) Jan. 2006 16.9 38 (3.4) 7.55 1,314 680 3 Nov. 2005 17.6 76 (7.1) 6.40 411 211 3 Jan. 2006 19.6 32 (3.7) 6.92 590 290

The neighbouring Vasse and Sabina catchments to the south show similarly high nutrient levels with TP and TN levels frequently exceeding ANZECC/ARMCANZ (2000a) guidelines (Scott 2000).

3.3 Macroinvertebrates 3.3.1 Taxonomy and Species Richness In total, 111 taxa were recorded from the eight Ludlow/Tiger Gully sites sampled in November 2005 with an additional 24 taxa recorded from the Capel River in January 2006. This represented a moderately diverse fauna for seasonal and perennial waters, respectively. Most taxa recorded were considered tolerant of a wide range of environmental conditions, common and frequently encountered in both seasonal and perennial river systems in south-western Australia. The fauna was dominated by insects (74%) with at least 62 species representing 18 families. Chironomidae (non-biting midges)

______Wetland Research & Management ______18 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06 constituted 24% of all insect taxa. Crustacea (including freshwater crayfish) constituted only 9% of the total fauna. A list of all taxa collected is presented in Appendix 4, indicating which taxa are likely to be ‘permanent’ residents and which are likely to be ‘temporary’, rapid colonisers following rain. All insects recorded, were considered to be ‘temporary’ residents with highly mobile adult phases that would allow them to avoid adverse environmental conditions (e.g. seasonal drying) and reinvade from nearby permanent waterbodies, once conditions have improved. The other 26% of fauna were considered ‘permanent’ residents (e.g. molluscs, crustaceans, annelids), which over-summer by burrowing into moist bottom and/or bank sediments or have modifications to avoid desiccation (i.e. water-tight seals on shells).

The taxonomic listing also includes records of larval and pupal stages for groups such as Diptera (two- winged flies) and Coleoptera (aquatic beetles). Current taxonomy in Australia is not sufficiently well developed to allow identification of all members of these groups to species level. In many instances it is likely that the pupal stages are the same species as the larval/adult stages recorded from the same location. However, because this could not be definitively determined, they were treated as separate taxa. Table 6 provides a summary of the major types of invertebrate fauna collected. Spatial variation in species richness was high with the number of species at individual sites in the Ludlow system ranging from 25 (upper control site 3) to 69 (conservation area, site 7) (Figure 5). This compared with 59 taxa recorded from the Capel River site in January 2006. Iluka’s rehabilitation site, Site 4 (TGD) had a comparatively good diversity with 49 taxa recorded. Across all sites, numerically dominant groups included the Diptera with 33 species from nine families, followed by Coleoptera (aquatic beetles) with 27 species from two families, and Crustacea with 10 species from at least five families. The higher diversity recorded from Site 7 was due to the presence of mites, amphipods and numerous beetle species. There appeared to be only limited seasonal variation in species richness in the Ludlow River, based on limited re-sampling (sites 2 & 3 only) in January 2006 concurrent with Capel River sampling. The average species richness recorded from Ludlow River in January 2006 was 44, compared to 40 in November 2005.

Table 6. Summary of invertebrate fauna collected 80 from the Cloverdale study area. 70 Macroinvertebrates No. of 60 ‘species’ a

x 50 a

T Turbellaria (flat worm) 1

f 40 o

. Nematoda (round worm) 1+

o 30 N 20 Temnocephala 1 10 Oligochaeta (aquatic worms) 1+ 0 Hirudinea (leeches) 2 1 2 3 4 5 6 7 8 9 Mollusca (snails & bivalves) 3 Site Acarina (water mites) 11

Figure 5. Species richness across sites Micro-crustacea 3+ (seed shrimps, water fleas) surveyed in the Ludlow system in Nov. 2005 and in the Capel River (Site 9) in Jan. 2006. Macro-crustacea (crayfish & amphipods) 7 Collembola (spring-tails) 2+ Diptera (two-winged flies) 42 Odonata (dragonflies & damselflies) 3 Trichoptera (caddis-flies) 9 Ephemeroptera (mayflies) 4 Hemiptera (true bugs) 13 Coleoptera (aquatic beetles) 32 Total number of ‘species’ 135

______Wetland Research & Management ______19 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Abundant at 80% of sites were: Chironomidae (non-biting midge larvae), Copepoda, Ostracoda (seed shrimps), Simulidae (black fly midge larvae), Corixidae (water boatmen), the notonectid Anisops sp. and the dytiscids (diving beetles) Rhantus suturalis and Platynectes desempunctatis (var. polygrammus). Other common taxa included oligochaetes (aquatic worms), the planorbid snail Glyptophysa (Glyptophysa) sp. and dytsicids Limbodessus inornatus and Necterosoma darwini. The amphipods Perthia branchialis and Austrochiltonia subtenius were common at downstream sites in the Ludlow River, but were not recorded upstream. ‘Low-occurrence’ taxa (i.e. taxa recorded from ≤10% of sites) accounted for nearly one third of all species and included Ephemeroptera (mayflies), Zygoptera (damselflies), a number of diptera and the amphipod Perthia acutitleson.

Taxa that were only recorded from the Capel River in January 2006 included the freshwater mussel Westralunio carteri, the freshwater shrimp Palaemonetes australis, eight chironomids, five beetle species, the caddis-flies Taschorema pallescens and Smicrophylax australis, at least one species of leptophlebiid mayfly and the dragonflies Austroaeschna anacantha and Hemicordulia sp. These species were not recorded in the Ludlow River neither in November 2005 nor when sites 2 (YWLRU) and 3 were re-sampled in January 2006, concurrent with Capel River sampling. Conversely, several taxa were only recorded from Ludlow/Tiger Gully sites and these included the freashwater snail Glyptophysa (Glyptophysa) sp., the chironomids Corynoneura sp., Limnophes pullulus and Procladius villosimanus, the corixids Micronecta robusta, Agraptocorixa parvipunctata, the notonectid Anisops and the dytiscid beetles Antiporus, Platynectes and Rhantus.

These differences reflect seasonal variations in community structure and availability of in-stream habitats in turn influenced by the differing flow regimes (permanent vs seasonal flow). For example, the chironomids Procladius villosimanus and Corynoneura sp. which were only present in the Ludlow system are known to be a primarily lentic (still-water) species. Planorbid snails such as Glyptophysa, corixids and most dytiscids typically inhabit lentic waters and slow flowing or stagnant areas of streams (Anderson & Weir 2004) and many are tolerant of poor water quality and degraded conditions.

Species generally known from lotic environments and requiring flowing waters include the caddis-flies Taschorema pallescens and Smicrophylax australis. They depend on the constant flow of water for delivery of oxygen, attaching themselves to rocks to spin nets for catching prey and plant material as its swept past. They are not common in heavily polluted or stagnant waters.

The amount of shelter and food afforded by bankside and in-stream vegetation/fallen debris will also determine taxa present, typically with greater diversity at sites with more remnant vegetation. The freshwater shrimp Paleomenetes australis, which was only present in the Capel River, is typically associated with snags, trailing vegetation and aquatic macrophytes (Gooderham & Tsyrlin 2002). Many leptophlebiid mayflies also prefer waters with a high abundance of wood and aquatic plants where they feed on algae and detritus (Gooderham & Tsyrlin 2002).

A greater diversity and abundance of caddis-flies, mayflies, dragonflies and damselflies is usually taken as indicative of improved water quality and overall better stream health. While the greater diversity of these species present in the Capel River may reflect generally better conditions, it is also likely a function of permanent flows. It must also be noted that while the majority of these species require good water quality and tend to be used as biological indicators, a number of taxa are known to be pollution-tolerant, including Tasmanocoenis tillyardi which was present in the Capel River.

3.3.2 Conservation Significance of M acroinvertebrates Twenty-five taxa (18%) recorded were considered south-west endemics, but only one was considered to have a restricted distribution within the south-west. This was the freshwater mussel Westralunio carteri, which was only recorded from the Capel River. The mussel is currently listed as a Priority 4 species under the CALM’s Wildlife Conservation (Specially Protected Fauna) Notice 2005 and as ‘vulnerable’

______Wetland Research & Management ______20 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06 under the IUCN Red List of Threatened Species (2004). Priority 4 species are those not currently threatened, but with fragmented and/or potentially vulnerable populations in need of monitoring. CALM uses IUCN categorises to define the conservation status of species at State, national and international levels. Criteria for the mussels’ IUCN listing include: ° a decline in area of occupancy, extent of occurrence and/or quality of habitat and ° its geographic distribution is precarious for the survival of the species and is limited, with the extent of occurrence estimated to be less than 20,000 km2.

Though Westralunio carteri may be locally common in some areas, many populations are in decline due to secondary salinisation and heavy siltation. W. carteri is a filter feeder and vulnerable to water pollutants and sedimentation. It is a known bio-accumulator of heavy metals and has often been used as such in bio-monitoring programmes. It is also intolerant of high salinity and levels above 4,000 µS/cm (~2,500 mg/L) may prove fatal. The mussel prefers shallow water habitats with stable, sandy or muddy bottoms and inhabits both permanent and seasonal rivers. It can survive prolonged periods of drought by burrowing into bottom muds and sealing the bivalve. It may thus survive potential drawdown of river pools associated with mine dewatering.

The hydropsychid caddis-fly Smicrophylax australis which was also only present in the Capel River is often reported in the literature as having a limited south-west distribution, known only from fragmented populations. However, Bunn 1988, Bunn et al. 1986 and Dean and Bunn 1989 report it as one of the most abundant and widespread caddis-flies in small perennial streams of the northern jarrah forest. This disparity may be due to a lack of detailed surveys of such smaller streams in recent years. Though it may be locally and seasonally abundant, a precautionary approach should be adopted toward its conservation until more about its biology is understood. It appears to require permanently flowing water and has rarely been recorded from heavily degraded systems.

Other south-west endemics collected during the current study included three freshwater crayfish (marron, gilgies & koonacs) and the amphipods Austrochiltonia subtenius and Perthia species (refer Appendix 4 for distributions across sites). The endemicity of the remaining 41% of taxa was indeterminate owing either to taxonomic resolution or lack of published data on statewide distributions. A conservation category code pertaining to endemicity was ascribed to each taxon and is provided in Appendix 4. Sites were ranked by percentage contribution of south-west endemics to the total recorded from each (Table 7). The Capel River (Site 9) had the highest proportion with 24% and Site 4 the lowest with 8%. The greater proportion of endemics supported by the Capel River was considered primarily due to the perennial flows and to greater heterogeneity of in-stream habitats in the form of large woody debris. There was no significant difference in the number of endemics between control and exposed groups along the Ludlow River/Tiger Gully (one-tail t-test assuming unequal variances, df = 4, t-stat =-0.471, p = 0.327).

Table 7. Sites ranked by number of SW endemics as a proportion of the total taxa recorded from each.

Site Rank % of total No. of SW endemics 9 (Capel) 2 24 14 8 1 19 8 3 3 16 4 2 4 15 6 5 5 14 5 1 6 13 7 6 7 12 5 7 8 12 8 4 9 8 4

______Wetland Research & Management ______21 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Only one introduced (exotic) species was recorded, the black-fly Simulium ornatipes which was present at most sites, except control sites 2 and 3. This species has become naturalized throughout much of the state, but is believed to have been originally introduced from the northern hemisphere (Williams 1980). Larvae were often collected in large numbers from rocks in riffle zones and water falls. Adult blackflies were present in large swarms at many of the sites surveyed.

3.3.3 Functional Feeding Groups The functional complexity and ‘health’ of a river system is influenced by the diversity of functional feeding groups4 (Cummins 1974, Cummins & Klugg 1979 (the obligate feeding mode of each species). Functional feeding groups were assigned to each species for which information could be gleaned from the literature (Figure 6a). Sources included Williams 1980, Barnes 1987, Bunn 1988, Boulton 1989, Cartwright 1997, Davis and Christidis 1997, St Clair 2000, Gooderham and Tsyrlin 2002 and the UWA database. The proportions of each functional feeding group from all eight sites are presented in Figure 6a. The system was found to have a high number of predators and collectors, with very few grazers or shredders.

(a) 100% 90% 80% Other/unknow n 70% Predators 60% Grazers 50% Filterers 40% Shredders 30% Collectors 20% 10% 0% 1 2 3 4 5 6 7 8 9 Site

(b)

control sites exposed sites

Collectors Shredders Filterers Grazers Predators Other/unknow n

Figure 6. Proportional occurrence of functional feeding groups of macroinvertebrates collected during from the Ludlow system in November 2005 (n=111) and from the Capel River in January 2006 (n=59); (a) histogram of all sites individually and (b) pie-charts of Ludlow/Tiger Gully control group (sites 1, 2, 3 & 4) vs exposed group (sites 5, 6, 7 & 8).

4 Functional feeding groups: ‘shredders’ feed on coarse particulate matter (CPOM >1mm); ‘collector’s feed on fine particulate matter (FPOM < 1mm); ‘filterers’ filter suspended particles from the water column and are often viewed as a subset of collectors; ‘grazers’ are those animals that graze or scrape algae and diatoms attached to the substrate; ‘predators’ capture live prey.

______Wetland Research & Management ______22 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Coleopteran and hemipteran species dominated the predators, while collectors were comprised of a variety of invertebrate taxa. The relative proportions of each functional feeding group were very similar between control and exposed groups (Figure 6b). For future monitoring, changes in feeding groups within exposed sites relative to control sites may indicate mine-related effects on community structure.

Current theories of functional organization of streams in the south-west (see Bunn 1985, 1986, 1988) predict relatively undisturbed, forested upland streams to be similarly dominated by collectors and predators, but with a high proportion of shredders. Proportions of each group are also expected to change seasonally and longitudinally within the catchment. Shredders would be expected to decrease downstream as the input of coarse particulate material decreased, while collectors and grazers would increase. The paucity of shredders within the current study area was not surprising, given the degree of riparian vegetation clearing with few overhanging, fringing or aquatic plants and hence reduced coarse particulate matter on which shredders feed. Collectors are also likely to dominate lower reaches of rivers and in particular disturbed reaches where the input of fine particulate material is high. In disturbed systems such as the Ludlow, an increase in grazers would also be expected as nutrient enrichment and increased light and higher water temperatures (due to a more open vegetation canopy) are likely to promote algal growth. However this did not appear to be the case, possibly due to seasonality of flows limiting permanent habitat for grazers and/or to unstable bed materials within the channels restricting algal and macrophyte growth and hence reducing availability of food sources for grazers.

3.3.4 Patterns in M acroinvertebrate Community Structure Though species richness varied between individual sites, there were no significant differences (ANOVA; df = 1, F = 1.315, p = 0.2951) between upstream controls (Sites 1, 2, 3 & 4) for Cloverdale and downstream exposed sites (Sites 5, 6, 7 & 8). Nor was there any difference between controls outside the influence of existing Yoganup and Yoganup West mines (i.e. sites 2 & 3) and control sites downstream from Yoganup West (sites 4 & 5) (one-tail t-test assuming unequal variances, df = 2, t value = -0.847, p = 0.243). Therefore sites selected as controls for Cloverdale should serve as good pre-mine reference sites against which to compare any future changes in macroinvertebrate species richness of the receiving environment.

Cluster analysis on macroinvertebrate abundance data separated the nine sites into five main groups (Figure 7). Not unexpectedly, community structure in the Capel River in January 2006 (group 5, Figure 7) was clearly distinct from that at the Ludlow/Tiger Gully sites sampled in November 2005. The four Cloverdale exposed sites (group 1) separated from most of the control sites with the exception of Site 4 (YOD) which grouped with exposed sites (sites 5, 6, 7 & 8). Cluster analysis isolated control sites 1, 2 and 3 as comparatively dissimilar to each other (Figure 7). When superimposed on the MDS, only the Capel River (group 5) and Ludlow River control site 3 (group 4) showed clear separation in ordination space (Figure 8), with all other sites grouping together due to the overriding influence of these two sites on the analysis. The low species diversity at site 3 separated it from all other sites (see further comment below). When the MDS ordination was re-run without Capel River (Figure 9a) and with Site 3 also excluded as an outlier (Figure 9b), there were only ‘loose’ groupings of sites that broadly reflected those from the cluster analysis, but no distinct patterns.

In general, the ordination of macroinvertebrates on presence/absence data demonstrated very similar patterns to abundance data, with comparable group membership and comparable position of samples in ordination space with no strong groupings of sites apparent. This is a good result with respect to detecting future mine-related effects, whereby future separation of sites will be apparent should control sites separate from exposed.

______Wetland Research & Management ______23 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

8

7

4

6

5

Group 1 S E T I S

Group 2 2

1 Group 3

3 Group 4

Group 5 9

20 25 30 35 40 45 50 55 60 65 70 75 80 85 90 95 100

SIMILARITY

Figure 7. Cluster dendrogram using macroinvertebrate log10 abundance data, indicating four main groupings.

Stress: 0.01 3

5 1 64 872

9

Figure 8. MDS ordinations using macroinvertebrate log10 abundance data for all sites with sites labelled by site codes and coloured by a priori groupings: p Ludlow/Tiger Gully control sites; r Ludlow/Tiger Gully controls for Cloverdale but exposed to existing Yoganup & Yoganup West; p Ludlow/Tiger Gully exposed sites for Cloverdale, Yoganup West and Yoganup; ø Capel River, January 2006.

______Wetland Research & Management ______24 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

(a) All Ludlow/Tiger Gully sites Stress 0.01

(b) Ludlow/Tiger Gully excluding Site 3 Stress 0.08 7

1 8

6

2

4

5

Figure 9. MDS ordinations using Ludlow/Tiger Gully macroinvertebrate log10 abundance data for a) all sites and b) excluding site 3 with sites labelled by site codes and coloured by a priori groupings: p control sites,r controls for Cloverdale but exposed to existing Yoganup & Yoganup West, p exposed sites for Cloverdale, Yoganup West and Yoganup.

Pairwise comparisons of groups using ANOSIM detected no significant separation between any combinations of groups; neither for a priori combinations (e.g. controls vs exposed) nor a posteriori groupings from the cluster analysis. Thus there appeared to be no difference in macroinvertebrate fauna composition between controls (sites 2 & 3) and those exposed to existing Yoganup/Yoganup West mining (sites 1 & 4) or future exposed sites for Cloverdale (sites 5, 6, 7 & 8). However, the small number of replicate sites within control and exposed groups likely reduces the statistical power of the analyses, with variation within the groups possibly overwhelming any statistically significant differences between the groups.

Between-site pairwise similarity values for the Ludlow River and Tiger Gully were tabulated to illustrate levels of site similarity (Table 8). Variation in abundance of macroinvertebrate fauna between sites was low to moderate, with pairwise combinations showing 34 – 67% similarity. Results indicated lowest percentage similarity for pairwise combinations with Site 3. The highest similarity was between sites 7 and 8 (66.67%). Mean between-site similarity for the control group (sites 1, 2, 3 & 4) was only 43%

______Wetland Research & Management ______25 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06 compared to 60% within the exposed group (sites 5, 6, 7 & 8), indicative of significantly greater faunal diversity between control sites than between exposed sites (one-tail t-test assuming unequal variances, df = 8, t value = -3.684, p = 0.0031).

Typically, between-site pairwise similarity was slightly greater using presence/absence than abundance data.

Table 8. Matrix of association values indicating percent pairwise similarity amongst the eight sites, calculated using the Bray-Curtis association measure on macroinvertebrate log10 abundance data.

Site 1 Site 2 Site 3 Site 4 Site 5 Site 6 Site 7 Site 8 Site 1 0 0 0 0 0 0 0 0 Site 2 49.16201 0 0 0 0 0 0 0 Site 3 34.89933 36.06557 0 0 0 0 0 0 Site 4 53.60825 52.69461 32.11679 0 0 0 0 0 Site 5 54.54545 52.17391 44.44444 61.43791 0 0 0 0 Site 6 57.89474 55.21472 46.61654 59.55056 65.77181 0 0 0 Site 7 56.30252 54.9763 39.77901 61.06195 51.77665 61.26126 0 0 Site 8 50 60.35503 46.04317 61.95652 55.48387 60 66.66667 0

The MDS ordination and separation of sites into broad groups was influenced by variations in the abundance of numerous taxa each with a relatively low percentage contribution to the overall similarity (or dissimilarity) of the groups. No one taxa could be viewed as truly indicative of control or of exposed groups or of an individual site. Species contributing most to group separation were cyclopoid copepods, ostracods, the black fly larvae Simulium ornatipes, the midge larve Thienemanniella sp., Tanytarsus sp. 1, Chironomus aff. altermans, Dicrotendipes conjunctus and the beetles Rhantus suturalis and Platynectes spp. Examples of these are represented as ‘bubble’ plots in Figure 9. With the exception of copepods and ostracods (Figure 9a), most of these species tend to be tolerant of pollutants and degraded conditions. Site 3 differed from all other sites primarily due to the absence of trichopteran (caddis-flies) and hemipteran (true bugs) species and the presence of very few dytiscids (aquatic beetles).

The relationships between the macroinvertebrate fauna and physico-chemistry were investigated using the BIOENV routine within PRIMER. Patterns in physico-chemical variables that fitted best with macroinvertebrate groups from ordination analyses were higher Al, Cd, Ni, Mn, Ba, NH3, pH, salinity (TDS, Econd. & associated concentrations in K+ and Na+), channel width and channel depth. Examples of some of these are represented as ‘bubble’ plots in Figure 10. Al and Cd levels were noticeably greater at control Site 3 which had the lowest faunal diversity (Figure 10a). While levels were inferred to be similar at Site 2 (upstream from Site 3), they were not actually measured at Site 2, which had a much greater faunal diversity than Site 3. Site 4, downstream from Site 3 also had a greater biodiversity and lower levels of Al and Cd. Maximum Ni concentrations were recorded at Site 1 and these may be influencing the type of species present (Figure 10b). Again, condition and nature of the riverine landscape must also be considered as a determining factor; i.e. the shallow, slower-flowing wetland area surrounding Site 1 and narrow channel (Figure 10d) may influence community structure to an equal or greater extent than the level of Ni present.

______Wetland Research & Management ______26 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

(a) Ostracoda spp. (b) Simulium ornatipes Stress: 0.08 Stress: 0.08

7 7

1 8 1 8

6 6

2 2

4 4

5 5

(c) Chironomus aff. alternans (d) Rhantus suturalis Stress: 0.08 Stress: 0.08

7 7

1 8 1 8

6 6

2 2

4 4

5 5

Figure 10. Bubbles plots superimposed on the MDS ordination (excluding Site 3), showing examples of the variation in log10 abundance classes between sites for four macroinvertebrate taxa. Sites are labelled by site code and the bigger the’ bubble’, the greater the species abundance. Compare with Figure 10 for corresponding variations in physico- chemical parameters at each site.

(a) Cadmium (b) Nickel

Stress: 0.08 Stress: 0.08

7 7

1 8 1 8

6 6

2 2

4 4

5 5

(c) Ammonia (NH3) (d) Average channel width

Stress: 0.08 Stress: 0.08

7 7

1 8 1 8

6 6

2 2

4 4

5 5

Figure 11. Bubble plots superimposed on the MDS ordination (excluding Site 3), showing examples of physico- chemical variables that correlated best to patterns in macroinvertebrate log10 abundance classes. Sites are labelled by site code and the bigger the ‘bubble’ the greater the metal concentration (a, b & c) or width (d). ______Wetland Research & Management ______27 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

3.3.5 Comparisons with Other M acroinvertebrate Studies The taxa recorded during the current study were those expected from disturbed watercourses on the coastal plain. Taxa normally associated with less disturbed ecosystems were rare or absent (e.g. Plecoptera, Ephemeroptera, Odonata and many of the Trichoptera). There is little published data on aquatic fauna of the study area, though the nearby artificial Capel Wetlands (Cale & Edward 1993) west of Bussell Highway have been extensively surveyed. Many of the insect taxa recorded from Cloverdale are also known to occur in these wetlands. The average species diversity in the Ludlow/Tiger Gully system was 40 and similar to the spring maxima reported for the Capel and associated natural wetlands such as Higgins Pond (Cale & Edward 1990, 1991a,b, 1993, 1994). Direct comparisons between wetland and river fauna, however should not strictly be made as lentic and lotic waterbodies typically support different faunal assemblages.

Comparisons with recent studies of other rural, coastal plain rivers, showed the Ludlow to support a macroinvertebrate fauna most similar to Henty Brook in the Collie River catchment (WRM 2006) which was sampled concurrent to the Ludlow. Many taxa were also shared in common with Mcknoe, Wellesley and Samson brooks (Creagh et al. 2004) and upper tributaries of Mayfield Drain (WRM 2005) – all in the Harvey River catchment to the north and all sampled during spring. Like the Ludlow, aquatic fauna in all these other seasonal systems was dominated by predators and collectors with a high proportion of Coleoptera and Diptera and a total taxa of around 90 – 100 (cf 92 in the Ludlow/Tiger Gully system, with taxonomy standardised between the various studies). The Ludlow supported far more taxa than heavily degraded systems such as Waroona/Drakesbrook Main drains with only 49 ‘species’ (WRM 2003).

3.4 Fish and Crayfish Three species of native freshwater crayfish and four species of fish were collected from the study area (Table 9). Local landowner Tom Hutton, reported another three species as present in the Capel River – cobbler, redfin perch and pouched lampreys. As sampling was not quantitative, results have been tabulated as relative abundance categories rather than total counts per site. All native species are common and widely distributed in south-west rivers. The western pygmy perch, western minnow and nightfish are probably the most common native fish in both seasonal and perennial rivers. Many pygmy perch caught were displaying breeding colours indicating the river does provide suitable breeding habitat for this species and likely also for western minnows and nightfish.

The majority of native fish require permanent water, only colonising seasonal streams from adjacent permanent waters during wet season flows or residing in permanent ‘refuge’ pools during summer. Gilgies are also typical of both seasonal and perennial waterbodies, while koonacs tend to be associated with seasonal systems in particular inland swamps and lakes. Although gilgies and koonacs are capable of burrowing to avoid summer drying, soils must be moist to ensure their gills remain hydrated. Unlike gilgies and koonacs, marron require permanent water.

Published lists of known fish occurrences in the region are limited. There is anecdotal evidence (Tom Hutton, Capel, pers. comm.) that cobbler numbers in the Capel River have declined over the past 40 years and that this may be related, at least in part, to a reduction in weed beds and an associated reduced abundance of freshwater shrimp (Palaemonetes australis). The introduction of perch is also implicated in the decline of the native western pygmy perch on which they predate (T. Hutton, pers. comm.). Morgan et al. (1998) surveyed the lower Ludlow River and wetlands west of Bussell Highway, the Capel River at Capel township and Capel River South, east of the Cloverdale project area. The survey was part of a larger study of south-western fish conducted between 1994 and 1996 and incorporated museum records. Species additional to those recorded during the current study included the Swan River goby (Pseudogobius olorum) and redfin perch from the Ludlow River. Morgan et al. (1998) also collected ammocoets (larvae) of the pouched lamprey (Geotricha australis) in low numbers from the Capel River and Capel River South. Both adult and larval lampreys are known to be seasonally

______Wetland Research & Management ______28 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06 common within the current study area (T. Hutton, pers. comm.), often present in irrigation pumps and channels. The larvae congregate in shallow sandy reaches.

Morgan and Beatty (2004) intensively sampled the Vasse River and Vasse River Diversion Drain in 2003 - 2004. The Vasse system lies approximately 20 km south of the Cloverdale project area. Morgan and Beatty (2004) included points all along the river, from the estuary to the far upper catchment. Species in common with their study of the Ludlow River included western minnows, pygmy perch, nightfish, Swan River goby and mosquitofish. Additional species were mud minnow (Galaxiella munda), western hardy head (Leptatherina wallacei), sea mullet (Mugil cephalus) and introduced goldfish (Carassius auratus). Of these, it is unlikely that sea mullet, an estuarine-marine species would penetrate inland from estuarine reaches to the Cloverdale project area. The western hardyhead is also more typically associated with estuarine and lower river reaches, though it may be found far inland in salinised rivers, e.g. the Blackwood River. The Swan River goby, however may indeed occur within the study area, as although it is an estuarine species it is known to swim upstream far into fresh waters. Gobies, hardyheads and sea mullet are all common in estuaries and coastal drainages between the mid-west and far south coasts.

Table 9. Fish and crayfish species in the Ludlow River / Tiger Gully and the Capel River.

Site Species 1 2 3 4 5 6 7 8 9 (Capel R.) Gilgie -- -- abundant ------present abundant Cherax quinquecarinatus Koonac ------common abundant ------Cherax plebejus Marron abundant Cherax cainii Indeterminate crayfish present ------present -- -- Cherax sp. (juveniles) Western pygmy perch -- present present -- common -- present -- present Edelia vittata Nightfish -- present present present common ------common Bostockia porosa Western Minnow -- present present -- common ------abundant Galaxia occidentalis Cobbler ------common Tandanus bostocki Pouched lampreys ------common Geotricha australis Mosquitofish -- abundant ------abundant abundant abundant -- Gambusia holbrooki Redfin Perch ------abundant Perca fluviatalis Indeterminate fish fry -- -- present ------

3.4.1 Conservation Significance of Fish and Crayfish Fauna All of the native fish and crayfish species recorded are south-west endemics, but none are rare or restricted in distribution. However, the recent discovery of the mud minnow in the Vasse River (Morgan & Beatty 2004) is of interest as this species has a highly restricted5 distribution. Though common (and occasionally abundant) in coastal rivers between the Goodga River (near Albany) and Margaret River, elsewhere it is only known from small disjunct populations near Gingin (Morgan et al. 1998, Allen et al. 2002). The mud minnow is believed to have disappeared from much of its range as a result of habitat degradation through rural development and predation by introduced fishes (e.g. redfin perch & trout). It is typically found in dark (coloured), acidic (pH 3.0 - 6.0) seasonal waters and has

5 Listed as restricted by the Australian Society for Fish Biology.

______Wetland Research & Management ______29 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06 never been recorded from cleared agricultural areas. It can aestivate in the bed of wetlands to survive summer dry periods. In the Vasse River, it was only recorded from forested headwater reaches (Morgan & Beatty 2004). It is highly unlikely that this species would occur within the Cloverdale project area, given the degree of vegetation clearing.

The pouched lamprey recorded from the Capel River belongs to an ancient lineage of jawless fishes whose morphology has remained largely unchanged for approximately 280 million years. Geotricha australis is the only surviving species in Australia, though six genera with 41 species are known to occur elsewhere (Allen et al. 2002). Though most abundant in river systems south of Margaret River, museum records show they were once present as far north as the Collie River. Habitat alteration is believed to have lead to their loss from many areas due to agricultural run-off and loss of silty beds in slow-flowing reaches (e.g. buried by sand), large dam construction and salinisation (ammocoetes are not tolerant of saline water). Ammocoetes spend 4 - 5 years in freshwaters, before metamorphosing and migrating to the sea. Adults remain in the open ocean for at least two years before returning to the rivers to spawn. Spawning is believed to take place in November. It is unlikely that pouched lampreys would use the seasonal Ludlow River as spawning habitat.

Redfin perch recorded from the Capel River are aggressive piscivores and can cause the decline of native fish populations through competition and predation. They also predate on invertebrates including freshwater crayfish. If caught, neither redfin perch nor mosquitofish should be returned live to the water. The release of redfin perch into natural waterways should be discouraged.

Marron have high conservation significance as an icon species6 (Nickoll & Horwitz 2000). Icon species are important ecologically because not only do they respond well to habitat restoration activities, but they also hold significant charismatic appeal. Generally, icon species are used to enhance public understanding of environmental issues. The elimination of marron from any area is considered to have significant conservation implications (Nickoll & Horwitz 2000). Marron are believed far more sensitive to environmental fluctuations than are gilgies or koonacs and require permanent water in order to survive (Morrissy 1980, Morrissy et al. 1984, Holdich & Lowery 1988).

3.5 Frogs Tadpoles of only one species the slender tree frog (Littoria adelaidensis) were collected in low numbers from Site 3. No adult frogs were heard calling along the Ludlow or Capel rivers during the current surveys, though WRM recorded Glauert’s froglet (Crinia glauerti) and the squelching froglet (C. insignifera) during concurrent surveys in nearby project areas at Elgin and Tutunup.

Preferred habitats of all species range from permanent stream pools and farm dams to ephemeral swamps, inundated road verges and the base of granite outcrops where run-off water collects. The froglets tend to be associated more with shallow swamps, marshes and seasonally damplands. Slender tree frogs generally prefer slow-flowing water bodies with dense fringing vegetation. Slender tree frogs breed from late spring to early summer, while froglets are typically winter breeders, though C. glauerti is believed to breed following any rain (Tyler at el. 2000). Though all species are currently widespread and abundant throughout the south-west, they are potentially threatened by over development (vegetation clearing) and all are susceptible to fungal infections (e.g. chytrid) that have caused high mortality in some south-west frog populations.

Bamford (2001) reported at least nine frog species known to occur in the general area (Table 10). None are rare or restricted in distribution. That only one species was recorded during the current survey may reflect lack of suitable habitat or the fact that surveys were limited.

6 A priority conservation grouping similar to keystone, umbrella and indicator species or groups.

______Wetland Research & Management ______30 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Table 10. Frog species listed by Bamford (2001) as present in the Ludlow area.

Species Common Name Myobatrachidae (ground frogs) Crinia georgiana Quacking froglet Crinia glauerti Glauert’s froglet Crinia insignifera Squelching froglet Geocrinia leai Green-bellied froglet Heleioporus eyrei Moaning frog Limnodynastes dorsalis Pobblebonk frog Pseudophryne guentheri Guenther’s toadlet

Hyliidae (tree frogs) Litoria adelaidensis Slender tree frog Litoria moorei Motorbike frog

3.6 Turtles Turtles were not specifically targeted during the current surveys and none were sighted. However, western long-necked turtles (Chelodina oblonga) are known to be common in the Capel River (Tom Hutton, Capel, pers. comm.).

Long-necked turtles inhabit both permanent and seasonal waterbodies and have a wide distribution throughout the south-west. Though they can survive seasonal drying they will preferentially use permanent pools where available. They can migrate relatively long distances overland if local conditions deteriorate (Cogger 1986) or they can burrow into the sediment and aestivate. The turtles only eat when open water is present and diet includes tadpoles, fish, and invertebrates. In permanent waters, this species has two nesting periods (September-October and December-January) but in seasonal systems, nesting will only occur in spring. Turtles generally nest in moist sandy soils, often hundreds of metres from the rivers edge, and eggs take up to two hundred days to hatch.

3.7 Waterbirds Waterbird abundances were very variable across sites. It was not possible to survey the system over one day, and therefore it was not known if birds observed on successive days were the same birds observed the previous day or birds new to the system. Therefore, it was not appropriate to provide total counts of species at each site. Instead, presence/absence of each species seen in different parts of the system are summarised in Table 10.

Waterbird species observed during fieldwork included Pacific black duck (18 individuals in total), grey teal (4), Australian wood duck (20), straw-necked ibis (4), white-necked heron (1), white faced heron (2) and black-fronted dotterel (2). In addition to waterbirds seen, flocks of straw-necked ibis were observed in paddocks adjacent to the river. While the Vasse-Wonnerup wetlands are known to support large waterbird populations (McAlpine et al. 1989, Pen 1997, Bamford 2001) the narrow river channel and lack of a large, open water body means the Ludlow River is unlikely to support an abundance of waterbirds. Loss of fringing vegetation has reduced suitable nesting sites across the floodplain and nearby wetlands. Waterbird use tends to be highest in large permanent wetlands with high diversity and abundance of vegetation (Broom & Jarman 1983, Halse et al. 1993, Storey 1993, Balla 1994).

All species reported are common. None are listed under JAMBA/CAMBA treaties.

______Wetland Research & Management ______31 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Table 11. Waterbird species recorded from the study area.

Site Species 1 2 3 4 5 6 7 8 9 Black-fronted dotterel P Charadrius melanops Grey teal P Anas gibberifrons Pacific black duck P P P Anas superciliosa Australian wood duck P Chenonetta jubata Straw-necked ibis P Threskiornis spinicollis White faced heron P Ardea novaehollandiae White-necked heron P Ardea pacifica

3.8 Water Rats Though not recorded during the current study, native water rats Hydromys chrysogaster are known to be present along the Capel River within the study area (Tom Hutton, Capel, pers. comm.). Water rats are common around coastal Australia and New Guinea, occurring in a wide range of coastal, brackish and freshwater environments (Watts & Aslin 1981). Habitats include rivers, swamps, lakes and drainage channels. They build nests at the ends of tunnels dug into banks near tree roots or in hollow logs. Breeding can occur throughout the year, but more typically in spring. They are omnivorous, feeding on crayfish, mussels, fish, plants, water beetles, dragonflies and smaller mammals and birds. Major threats to the species are predation by cats, loss of habitat through clearing and grazing and loss of aquatic food sources due to secondary salinisation (Lee 1995). Despite this, many populations thrive in highly urban areas. They suffer from heat stress if access to permanent water is lost.

4 CONCLUSIONS 4.1 Ecological Values In general, the riverine ecosystems of Ludlow River and Tiger Gully were considered of limited regional conservation value due to past anthropogenic disturbances, including drain construction, disturbance of the riparian zones (livestock, weed infested and erosion prone) and loss of in-stream habitat. While drain construction has significantly reduced the size of natural seasonal wetlands, damplands and palusplains, clearing has resulted in increased overland flows and rising water tables leading to likely higher than historic flows within the Ludlow/Tiger Gully channels. Aquatic fauna communities at all sites were dominated by cosmopolitan species, typical of lowland rural regions. Though waterbird species may visit the river, its seasonal nature, the small size of the channels and lack of suitable vegetation were considered unlikely to provide significant dry-season habitat or breeding habitat.

The Capel River, Ludlow River and Tiger Gully do however support freshwater crayfish (marron, gilgies & koonacs) of local conservation significance. The Capel River and designated conservation area along the Ludlow River, downstream from the project area, currently supports healthy stands of remnant overstorey vegetation with evidence of strong recruitment. Macroinvertebrate biodiversity also tended to be greatest in these areas. These factors confer relatively high conservation significance to these reaches. The Capel River was also considered of high ecological value due to the greater number of endemic and indicator species it supports and to the presence of the Priority 4 freshwater mussel Westralunio carteri.

______Wetland Research & Management ______32 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Eucalypt vigour within the Ludlow conservation area and along the Capel River may provide a useful indicator of any detrimental changes to flow regime as a result of either channel diversion along the Ludlow or pit de-watering (e.g. changes to depth or duration of saturation of alluvial soils) within the riparian zone.

4.2 Potential Threats from Mine Activities While current mine operations at Yoganup and Yoganup West have had no apparent adverse effect on aquatic fauna communities, there is the suspicion that mining may be affecting water quality due to increased salinity, water hardness, alkalinity, sulphate, Zn and Ni. For the metals at least, this effect would appear to be highly localised. Spot measurements of water quality parameters are unlikely to reflect total loads. Nor has bioavailablity of elevated metals been assessed. Sub-lethal (chronic) effects on fauna are also difficult to detect without laboratory toxicity testing on target species. The existing mines have not been in operation long and though the life of each and of the Cloverdale mine is expected to be short (2-3 years), their cumulative effect on the downstream receiving environments should be monitored.

The coastal plain is known to have areas of high acid sulphate soil risk, particularly low-lying wetlands and damplands. If exposed due to mining there is the direct threat of acid on the environment and the indirect threat of lowered pH mobilising metals, increasing their bioavailability and toxicity. If metals are ‘naturally’ elevated, the risk is even greater. Other threats posed by exposed acid sulphate soils are poor soil fertility, vegetation dieback, surface scalding and erosion.

It is recommended that local water quality guideline values and indices be derived from background levels as measured in current baseline assessments. The ANZECC/ARMCANZ (2000a) guidelines for general water quality parameters and for metals are conservative and as such do not necessarily reflect existing levels (naturally elevated or anthropogenically elevated) within the catchment. For example, Al, Cr and Zn appear already to be elevated at control sites. For any elevated concentrations associated with mining, consideration should be given to the deployment of DGTs to determine bio-available components, with levels referred against ANZECC guidelines. While fauna may have some level of tolerance/acclimation to naturally higher background levels, any further elevation due to mining could exceed thresholds of bioavailable metals and result in lethal or sub-lethal toxic effects.

While the current proposal for Cloverdale is unlikely to directly affect surface flows, there is the potential for indirect impacts through groundwater drawdown, further vegetation-loss, increased soil nutrients and contaminants and the release of mine dewatering discharge into ‘natural’ watercourses. Since the Ludlow system is naturally seasonal, maintenance of summer surface flows would not be required to maintain the existing riverine ecosystems. However survival of remnant vegetation and some invertebrate fauna (e.g. gilgies and koonacs) will be dependent on depth and spatial extent to which water tables are lowered over summer and the impact to soil moisture content at depth. Similarly, most macroinvertebrates fulfil their entire larval stage within an aquatic environment and as such, may be made locally extinct by any reduction in groundwater flows, if channel pools and sediments dry before the adults emerge. Larger pools, such as those on the Capel River are likely to provide summer refuge for marron, fish, turtle and water rats and should be maintained. The spatial extent of dewatering drawdown should be determined relative to any summer refuge pools.

Determination of the extent to which the vegetation along the Capel River and Ludlow River conservation area was reliant on groundwater, channel water (including overbank flows) and/or sheet flow, was beyond the scope of the current study. The seasonal nature of flow in the Ludlow means vegetation here is more likely dependent on groundwater to sustain it over summer. Any reduction in overbank flows during winter would more than likely be compensated by sheet flooding in both the

______Wetland Research & Management ______33 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Capel and Ludlow sub-catchments. However perennial flows in the Capel River would already appear under threat from excessive abstraction for summer irrigation for agriculture. Flows downstream of the mine will need to be monitored to ensure adequate volumes of water are actually being maintained and to determine the contribution of catchment runoff to both surface flow and groundwater recharge. In the event that mining necessitated larger than expected de-watering discharge volumes, discharge should be managed to avoid excessive water velocities which may lead to increased bed and bank erosion, channel incision and downstream sedimentation.

If diversion of the Ludlow River were to be required, the new channel should be of adequate size, dimension and elevation to act as a ‘natural’ creekline, in the manner of Iluka’s existing streamlining programme below Yoganup West. Design features which should be considered include: ° heterogeneity in plan form (meanders as opposed to a straight channel) and longitudinal profile (pools and riffles), ° appropriate bank profiles to avoid a steep, incised channel, ° construction of benches upon which riparian vegetation can ultimately establish (or be established using propagation and planting) and ° sufficient conveyance capacity to cater for low frequency but high magnitude rainfall/run off events.

4.3 Recommendations 1. Environmental management programmes for the Cloverdale project should seek to prevent further degradation of those areas that retain some ecological value and to facilitate the possible future restoration of degraded areas. In particular the Ludlow River conservation area needs on- going monitoring to ensure protection of existing biodiversity. 2. Iluka has already formulated programmes for routine monitoring of ground and surface water quality and quantity as well as plans for drainage management. These monitoring programmes should be maintained for the life of the mine and data adequately evaluated in light of elevated salinity, water hardness, alkalinity, Zn and Ni at YDP. 3. Establish new trigger values using local baseline metals data as per ANZECC/ARMCANZ (2000a). If levels increase above this trigger value, deploy DGTs to determine bioavailable components of dissolved metals and refer this to world database (available on-line) on metals effects to determine fauna susceptibility. 4. Target water quality sampling to isolate source of elevated parameters at YDP – i.e. could Tiger Gully be the source? 5. Consult with external water analysis laboratories to ensure detection limits for analytes (e.g. copper and chromium) meet trigger values. While achievable, this usually incurs additional cost. 6. Though winter flows are now greater than would have occurred historically, the seasonality of the flow regime has been maintained. Effort should be made to ensure any deeper permanent pools in the Ludlow River be maintained in the face of any dewatering drawdown to provide a summer dry-season refuge for aquatic fauna. 7. To protect habitat for freshwater crayfish it is recommended that turbidity in the Ludlow River downstream from mining should not increase by more than 10% above the existing seasonal mean concentration (based on ANZECC/ARMCANZ 2000a guidelines for the protection of aquatic ecosystems). Crayfish may become locally threatened due to loss of breeding habitat, shelter and food resources if mining were to lead to excessive siltation/aggradation and permanently increase turbidity. 8. To supplement Iluka’s ground and surface water monitoring, aquatic ecosystem monitoring is recommended to confirm there are no detrimental impacts from mineral sands mining.

______Wetland Research & Management ______34 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Ecosystem monitoring would also assist in determining the success of streamlining projects currently being undertaken by Iluka. At least biennial monitoring of aquatic fauna (macroinvertebrates & fish) and associated physico-chemical parameters and riparian vegetation condition is recommended. For comparative purposes, future sampling should adopt methods consistent with those used in the current baseline study. 9. Consider including the monitoring of eucalypt vigour as an indicator of change in surface/groundwater flow in the Ludlow River conservation area and along the Capel River.

______Wetland Research & Management ______35 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

5 REFERENCES Allen G.R. (1989). “Freshwater Fishes of Australia”. Neptune City: T.F.H. Publications. Allen G.R., Midgley S.H. & Allen M. (2002). “Field Guide to the Freshwater Fishes of Australia”. Western Australian Museum, Perth WA. Andersen N.M. & Weir T.A. (2004). Australian Waterbugs: Their Biology and Identification (Hemiptera – Heteroptera, Gerromorpha & Nepomorpha). Entomonograph Volume 14. CSIRO Publishing, Collingwood Australia. ANZECC/ARMCANZ (2000a). Australian and New Zealand Guidelines for Fresh and Marine Water Quality. Australia and New Zealand Environment and Conservation Council and the Agriculture and Resource Management Council of Australia and New Zealand. Paper No. 4. Canberra. http://www.deh.gov.au/water/quality/nwqms/index.html ANZECC/ARMCANZ (2000b). Australian and New Zealand Guidelines for Water Quality Monitoring and Reporting. Australia and New Zealand Environment and Conservation Council and the Agriculture and Resource Management Council of Australia and New Zealand. Paper No. 7. Canberra. http://www.deh.gov.au/water/quality/nwqms/index.html Balla S. (1994). Wetlands of the Swan Coastal Plain Volume 1. Their Nature and Management. Water Authority of Western Australia and the Western Australian Department of Environmental Protection. Bamford (2001). Ludlow Mining Lease fauna Survey March/April 2001. Interrim report prepared by M.J. & A.R. Bamford, Consulting Ecologists, to Cable Sands Barnes R.D. (1987). “Invertebrate Zoology”. Saunders College Publishing, Philadelphia. Boulton A.J. (1989). Over-summering refuges of aquatic macroinvertebrates in two intermittent streams in central Victoria. Transactions of the Royal Society of Society of South Australia 113: 23-34. Bunn S.E. (1985). Community structure and functional organisation of streams of the northern jarrah forest, Western Australia. Unpublished PhD thesis, The University of Western Australia. Bunn S.E. (1986). Spatial and temporal variation in the macroinvertebrate fauna of streams of the northern jarrah forest, Western Australia: functional organisation. Freshwater Biology 16: 621-632. Bunn S.E. (1988). Processing of leaf litter in a northern jarrah forest stream, Western Australia: seasonal differences. Hydrobiologia 162: 201-210. Bunn, S.E., Edward, D.H. and Loneragan, N.R. (1986). Spatial and Temporal Variation in the Macro- Invertebrate Fauna of Streams of the Northern Jarrah Forest, Western Australia: Community Structure. Freshwater Biology 16:67-91. Broom L.S. & Jarman P.J. (1983). Waterbirds on natural and artificial waterbodies in the Namoi Valley, New South Wales. Emu 83: 99-104 Cale D.J. & Edward D.H.D. (1990). The Influence of aquatic Macroinvertebrate Communities in Swamphen Lake, Western Australia. Technical Report No. 10. Unpublished report by the Aquatic Research Laboratory, Dept. Zoology, UWA. October 1990. Cale D.J. & Edward D.H.D. (1991a). Accumulated Changes in the Macroinvertebrate Community, Twelve Months after the Addition of Phosphate Fertilizer to Cadjeput, Capel, Western Australia. Technical Report No. 12. Unpublished report by the Aquatic Research Laboratory, Dept. Zoology, UWA to AMC Wetlands Centre. September 1991. Cale D.J. & Edward D.H.D. (1991b). The Response of the Benthic Macroinvertebrate Community to the Addition of Straw in Gravel Pool, Capel, Western Australia. Technical Report No. 11. Unpublished report by the Aquatic Research Laboratory, Dept. Zoology, UWA to AMC Wetlands Centre. September 1991. Cale D.J. & Edward D.H.D. (1993). A Comparison of the Macroinvertebrate faunas of a natural Wetland and Two Man-made wetlands at the RGC Wetlands Centre Capel, Western Australia. Technical Report No. 18. Unpublished report by the Aquatic Research Laboratory, Dept. Zoology, UWA to RGC Wetlands Centre. April 1993.

______Wetland Research & Management ______36 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Cale D.J. & Edward D.H.D. (1994). The Effect of the Addition of Hay on the Biomass and Diversity of macroinvertebrates in Cadjeput Pool, Western Australia. Technical Report No. 24. Unpublished report by the Aquatic Research Laboratory, Dept. Zoology, UWA to RGC Wetland Centre. September 1994. Cartwright D.I. (1997). A Key to Species of Late Instar larvae of Australian Trichoptera (Families Dipseudopsidae, Glossosomatidae, Polycentropodidae, Pschyomyiidae, Ecnomidae, Philopotamidae and Tasimiidae). Melbourne Water, Western Treatment Plant, Werribee, Victoria. Clarke K.R. & Gorley R.N. (2001). Primer v5: User Manual/Tutorial, Primer E: Plymouth. Plymouth Marine Laboratory, Plymouth, UK. Cogger H.G. (1986). “Reptiles and Amphibians of Australia”. Reed Press Australia. Creagh S., Davies P.M., Chandler L. & Lynas J. (2004). Aquatic ecosystems of the Upper Harvey River Catchment. Samson Brook, McKnoe Brook, Wokalup Creek, Wellesley Creek and Harvey River: fauna monitoring surveys November 2003. Streamtec Report ST 11.1/03. Unpublished report to the Water Corporation, Western Australia. Cummins K.W. (1974). Structure and function of stream ecosystems. Bioscience 24: 631-641. Cummins K.W. & Klugg M.J. (1979). Feeding ecology of stream invertebrates. Annual Review of Ecology and Systematics 10: 147-172. Davis J.A., Rosich R.S., Bradley J.S., Growns J.E., Schmidt L.G. & Cheal F. (1993). Wetlands of the Swan Coastal Plain, Volume 6. Wetland Classification on the Basis of water Quality and Invertebrate Community Data Wetlands of the Swan Coastal Plain. Water Authority of Western Australia and Environmental Protection Authority, Perth. Dean J.C. & Bunn S.E. (1989). Larval descriptions of the Hydrobiosidae, Philopotamidae, Hydropsychidae and some Ecnomidae (Trichoptera) from south-western Australia, with notes on biology. Australian Journal of Marine and Freshwater Research 40: 631-643. Gooderham J. & Tsyrlin E. (2002). The Waterbug Book: a Guide to Freshwater Macroinvertebrates of Temperate Australia. CSIRO Publishing, Collingwood, Victoria Halse, S.A., Williams, M. R., Jaensch, R.P., & Lane, J. A. K. (1993). Wetland characteristics and waterbird use of wetlands in south-western Australia. Wildlife Research 20: 103-126. Hart B.T. & McKelvie I.D. (1986). Chemical Limnology in Australia. [In] P. De Decker & W.D. Williams (eds) “Limnology in Australia.” CSIRO/DR Junk Publishers. pp 3-32. Holdich D.M. & Lowery R.S. (eds). (1988). “Freshwater Crayfish – Biology, Management and Exploitation.” Croom Helm, London, 167-212 IUCN (2004). 2004 Red List of Threatened Species. International Union for Conservation of Nature and Natural Resources. http://www.redlist.org/ Lee A.K. (1995). “The Action Plan for Australian Rodents.” Australian Nature Conservation Agency, Canberra. McAlpine K.W., Spice J.F. & Humphries R. (1989). The Environmental Condition of the Vasse-Wonnerup Wetland System and a Discussion of Management Options. Western Australian Environmental Protection Authority Technical Series 31: 1-35. Morgan D.L. & Beatty S. (2004). Fish Fauna of the Vasse River and the Colonisation by Feral Goldfish Carassius auratus. Unpublished report by the Centre for Fish and Fisheries Research, Murdoch University, to Geographe Catchment Council- Geocatch – and Fishcare WA. September 2004. Morgan D.L., Gill H.S. & Potter I.C. (1998). Distribution, Identification and Biology of Freshwater Fishes in South-western Australia. Records of the Western Museum Suppl. 56: 1-97. Morrissy, N.M. 1980. Production of Marron in Western Australian Farm Dams. Fisheries Research Bulletin of Western Australia 24 (W.A. Dept of Fisheries and Wildlife), 1-80. Morrissy N.M., Caputi N. & House R.R. (1984). Tolerance of marron (Cherax tenuimanus) to hypoxia in relation to aquaculture. Aquaculture 41: 61-74. Nickoll R. & Horwitz P. (2001). Evaluating flagship species for ecosystem restoration; a case study involving freshwater crayfish and a river in south-western Australia. [In] D.Saunders, D.Craig & N.Mitchell (eds)

______Wetland Research & Management ______37 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

“Nature Conservation V. Nature Conservation in the Production Matrix.” Surrey Beatty and Sons, Chipping Norton, New South Wales. Pen L. (1997). A Systematic Overview of Environmental Values of the Wetlands, Rivers and Estuaries of the Busselton-Walpole Region. Report No WRAP 7. Water and Rivers Commission, Perth Pen L. & Scott M. (1995). Stream Foreshore Assessment in Farming Areas. Blackwood Catchment Co- ordinating Group. National Landcare Program. Department of Agriculture WA. June 1995. Scott M. (2000). Vasse River Action Plan. Geographe Catchment Council – Geocatch – and the Vasse- Wonnerup Land Conservation District Committee. St Clair R.M. (2000). Preliminary Keys for the Identification of Australian Caddisfly Larvae of the Family Leptoceridae. Comparative Research Centre for freshwater Ecology Identification Guide No. 27. Presented at the taxonomy workshop held at the Murray Darling Freshwater Research Centre, Albury. 8-9 February 2000. Storey A.W., Vervest R.M., Pearson G.B. & Halse S.A. (1993). Wetlands of the Swan Coastal Plain Volume 7. Waterbird Usage pf Wetlands on the Swan Coastal Plain. Environmental Protection Authority and Water Authority of Western Australia. Tyler M.J., Smith L.A. & Johnstone R.E. (2000). “Frogs of Western Australia.” Revised Edition. Western Australian Museum. Watts C.H.S. & Aslin H.J. (1981). “The Rodents of Australia.” Angus and Robertson, Sydney. White K. & Comer S. (1999). Capel River Action Plan. Geographe Catchment Council – Geocatch – and the Capel Land Conservation District Committee. Williams W.D. (1980). “Australian Freshwater Life: The Invertebrates of Australian Inland Waters.: The MacMillan Company of Australia Pty. Ltd., South Melbourne, Australia. WRC (1999). Planning and Management: Foreshore Condition Assessment in Farming Areas of South-west Western Australia. Water and Rivers Commission River restopration Report No. RR3. http://portal.environment.wa.gov.au/pls/portal/docs/PAGE/DOE_ADMIN/GUIDELINE_REPOSIT ORY/RR3.PDF WRM (2003). Preliminary Ecological Water Requirements for Drakesbrook. Unpublished report to Harvey Water. June 2003. WRM (2005). Waroona Mineral Sands Project: Baseline Aquatic Ecosystem Surveys and Preliminary Social Water Requirements. Unpublished report by Wetland Research & Management to Iluka Resources Ltd. April 2005. WRM (2006). Burekup Project: Baseline Aquatic Biology and Water Quality Study. Unpublished DRAFT report by Wetland Research & Management to Iluka Resources Ltd. January 2006.

______Wetland Research & Management ______38 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

APPENDICES

______Wetland Research & Management ______39 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06 Appendix 1. Summary of physico-chemical data from each site. General water quality parameters and nutrient concentrations Secchi Site No. Date Time DO pH m Temp NH3-N NO3-N TN P-SR TP Channel Wetted Depth m hrs % (mg/L) In situ (Lab.) ºC mg/L mg/L mg/L mg/L mg/L width m width m Ave. (Max.) 1 (YOD) 11/11/05 1045 75 (6.7) 6.14 (6.7) >depth 20.3 <0.01 0.01 0.26 <0.01 0.02 1.0 0.75 0.10 (1.4) 2 (YWLRU) 14/11/05 1200 118 (12.6) 6.91 -- >depth 19.8 ------12.0 6.0 0.55 (0.4) 3 11/11/05 1530 76 (7.1) 6.40 (6.7) >depth 17.6 0.03 0.19 0.2 <0.01 0.02 12.5 8.0 0.50 (1.0) 4 (TGD) 11/11/05 1330 104 (9.1) 7.32 (7.6) >depth 22.8 <0.01 0.11 0.4 <0.01 0.02 8.0 4.5 0.35 (0.9) 5 14/11/05 1400 103 (9.1) 7.36 -- >depth 21.9 ------3.5 2.0 0.30 (1.4) 6 14/11/05 1045 90 (9.0) 6.93 (7.2) >depth 17.6 0.03 0.15 0.33 <0.01 0.03 6.5 5.0 0.45 (0.85) 7 15/11/05 0950 95 (8.6) 6.88 -- >depth 19.6 ------8.0 5.0 0.35 (0.85) 8 15/11/05 1150 80 (7.0) 7.20 -- >depth 21.1 ------8.0 5.0 0.35 (0.8) 9 (Capel R) 25/01/06 1400 86 (7.4) 6.47 -- 1.0 20.3 ------19.0 4.5 0.40 (1.8)

Electrical conductivity and composition of major ions

Site No. Date Time ECond. µS/cm TDS mg/L TSS Alkalinity Hardness Ca Na K Mg HCO3 CO3 Cl SO4 hrs In situ (Lab.) In situ (Lab.) mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L 1 (YOD) 11/11/05 1045 819 (762) 419 (430) 6 15 130 16.8 94.4 5.9 20.9 18 <2 175 94.9 2 (YWLRU) 14/11/05 1200 461 -- 237 ------3 11/11/05 1530 411 (387) 211 (200) 1 5 35 1.8 56.8 2.9 7.5 6 <2 111 9.9 4 (TGD) 11/11/05 1330 463 (440) 238 (200) 2 10 55 3.6 76.7 4.6 11.2 12 <2 122 19.9 5 14/11/05 1400 481 -- 247 ------6 14/11/05 1045 509 (454) 262 (220) 6 15 46 3 64.1 4.3 9.3 18 <2 126 14.8 7 15/11/05 0950 528 -- 272 ------8 15/11/05 1150 503 -- 259 ------9 (Capel R) 25/01/06 1400 1,115 -- 560 ------

Metal concentrations Site No. Date Time Al As Ba B Cd Co Cr Cu Fe Hg Mn Mo Ni Pb hrs mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L 1 (YOD) 11/11/05 1045 <0.005 <0.001 0.08 <0.02 <0.0001 <0.005 <0.002 <0.005 0.079 <0.0005 0.013 <0.001 0.01 <0.0001 2 (YWLRU) 14/11/05 1200 ------3 11/11/05 1530 0.058 <0.001 0.041 <0.02 0.0003 <0.005 <0.002 <0.005 0.11 <0.0005 0.019 <0.001 0.002 <0.0001 4 (TGD) 11/11/05 1330 <0.005 <0.001 0.043 <0.02 <0.0001 <0.005 <0.002 <0.005 0.13 <0.0005 0.006 <0.001 0.001 <0.0001 5 14/11/05 1400 ------6 14/11/05 1045 <0.005 <0.001 0.084 <0.02 <0.0001 <0.005 <0.002 <0.005 0.25 <0.0005 0.008 <0.001 0.001 <0.0001 7 15/11/05 0950 ------8 15/11/05 1150 ------9 (Capel R) 25/01/06 1400 ------______Wetland Research & Management ______40 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Metal concentrations cont. and petroleum hydrocarbons Site No. Date Time Se V Zn Benzene Toluene Ethylben Xylene TotalBTE C6-C9 C10-C14 C15-C28 C29-C36 TPH hrs mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L mg/L 1 (YOD) 11/11/05 1045 <0.001 <0.005 0.063 <0.001 <0.001 <0.001 <0.002 <0.005 <0.025 <0.025 <0.1 <0.1 <0.25 2 (YWLRU) 14/11/05 1200 ------3 11/11/05 1530 <0.001 <0.005 0.041 <0.001 <0.001 <0.001 <0.002 <0.005 <0.025 <0.025 <0.1 <0.1 <0.25 4 (TGD) 11/11/05 1330 <0.001 <0.005 0.016 <0.001 <0.001 <0.001 <0.002 <0.005 <0.025 <0.025 <0.1 <0.1 <0.25 5 14/11/05 1400 ------6 14/11/05 1045 <0.001 <0.005 0.045 <0.001 <0.001 <0.001 <0.002 <0.005 <0.025 <0.025 <0.1 <0.1 <0.25 7 15/11/05 0950 ------8 15/11/05 1150 ------9 (Capel R) 25/01/06 1400 ------

______Wetland Research & Management ______41 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Appendix 2. ANZECC/ARMCANZ (2000a) guideline levels

(a) General Water Quality Parameters NB: these trigger values applicable to south-west WA. Upland streams refers to those above 150m altitude.

Table A2-1. Trigger values nutrients, dissolved oxygen and pH for the protection of aquatic ecosystems and for livestock drinking water, applicable to south-west Western Australia (TP = total phosphorus; FRP = filterable reactive phosphorus; TN = total nitrogen; NOx = total nitrates/nitrites; NH4+ = ammonium).

+ TP FRP TN NOx **NO3 **NO2 NH3 NH4 DO pH (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) (mg/L) % saturation2 Aquatic Ecosystem

Upland River1 0.02 0.01 0.45 0.2 NP NP 0.95 0.06 90 6.5 – 8.0

Lowland River1 0.065 0.04 1.2 0.15 NP NP 0.95 0.08 80 - 120 6.6 – 8.0

Lakes & Reservoirs 0.01 0.005 0.353 0.01 NP NP 0.95 0.01 90 6.5 – 8.0

Wetlands3 0.06 0.03 1.5 0.1 NP NP 0.95 0.04 90 - 120 7.0 – 8.54

Livestock Drinking NP NP NP NP 400 30 NP NP NP NP Water

** Where 1mg/L NO3-N = 4.43 mg/L NO3; 1 mg/L NO2-N = 3.29 mg/L NO2. NP = value not provided. 1 All values during base river flow not storm events. 2 Derived from daytime measurements; may vary diurnally and with depth; data loggers required to assess variability. 3 Elevated nutrients in highly coloured wetlands do not appear to stimulate algal growth. 4 In highly coloured wetlands, pH typically ranges 4.6 – 6.5. 5 General level for slightly-moderately disturbed ecosystems and not specifically formulated for south-west WA; figure may not protect species from chronic toxicity.

Table A2-2. Trigger values for conductivity and turbidity for the protection of aquatic ecosystems and for livestock drinking water applicable to south-west Western Australia.

ECond. (
S/cm) Explanatory notes Aquatic Ecosystems Upland & lowland rivers 120 - 300 Values will vary depending on geology. First flush after seasonal rain may result in temporarily high values.

Lakes, reservoirs & wetlands 300 – 1,500 Higher values will occur during summer when water levels are reduced due to evaporation. Values in WA wetlands may exceed 3,000
S/cm.

Livestock Drinking Water 3,731 For dairy cattle; equivalent to approx. 2,500 mg/L TDS

Turbid. (NTU) Aquatic Ecosystems Upland & lowland rivers 10 - 20

Lakes, reservoirs & wetlands 10 - 100 Shallow lakes and reservoirs may have higher turbidity naturally due to wind-induced re-suspension of sediments. Lakes & reservoirs in catchments with highly dispersible soils will have high turbidity. Wetlands vary greatly in turbidity depending on general catchment condition, recent flow events and water level in the wetland.

Livestock Drinking Water Not provided

______Wetland Research & Management ______42 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

(b) M etals and Hydrocarbons

Table A2-3. Trigger values for metals and hydrocarbons for the protection of aquatic ecosystems and for livestock drinking water. Values for slightly to moderately disturbed aquatic ecosystems (i.e. low risk) are shaded grey.

Compound Trigger Value for Freshwater (mg/L) Trigger value for livestock Level of protection (% species) Level of protection (low risk) 99 95 90 80 Aluminium pH >6.5 0.027 0.055 0.080 0.150 5 Aluminium pH <6.5 ID ID ID ID NP Arsenic (III) 0.001 0.024 0.094C 0.360C 0.5 – 5T (total As) Arsenic (V) 0.0008 0.013 0.042 0.140C NP Boron 0.09 0.37C 0.68C 1.3C 5 Cadmium H 0.00006 0.0002 0.0004 0.0008 0.01 Calcium NP NP NP NP 1000 Chromium (III) H ID ID ID ID 1 (total Cr) Chromium (VI) 0.00001 0.001C 0.006 0.040 NP Cobalt ID ID ID ID 1 Copper H 0.001 0.0014 0.0018C 0.0025C 1 (cattle) Iron ID ID ID ID Not sufficiently toxic Lead H 0.0010 0.0034 0.0056 0.0094C 0.1 Magnesium NP NP NP NP 2000 Manganese 1.2 1.9C 2.5C 3.6C Not sufficiently toxic Mercury (inorganic) B 0.00006 0.0006 0.0019C 0.0054C 0.002 Molybdenum ID ID ID ID 0.15 Nickel H 0.008 0.011 13 0.017 1 Selenium (total) B 0.005 0.011 0.018 0.034 0.02 Sulphate NP NP NP NP 1000 Vanadium ID ID ID ID ID Zinc H 0.0024 0.008C 0.015C 0.031C 20 Benzene 0.6 0.95 1.3 2.0 NP Toluene ID ID ID ID NP Ethylbenzene ID ID ID ID NP Xylene 0.14 0.2 0.25 0.34 NP Oils & Petroleum ID ID ID ID NP Hydrocarbons

ID = indeterminate; NS = value not provided. B Metals for which bioaccumulation and secondary poisoning effects should be considered; C Level may not protect key test species from chronic toxicity; H Metals for which levels should be adjusted if water hardness (as CaCO3) >30mg/L - refer Tables A2-4 and A2-5 below. T Total arsenic; concentrations of 5.0 mg/L may be tolerated if not provided as food additive and natural levels in diet are low.

______Wetland Research & Management ______43 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Table A2-4. Approximate factors to apply to trigger values (TV) in Table A2-3 for freshwaters of varying hardness. If water hardness is away from the mid-range, it may be preferable to use the algorithm in Table A2-5.

Hardness category Mid-Range Cd Cr(III) Cu Pb Ni Zn Hardness

(mg CaCO3/L) (mg CaCO3/L) Soft (0-59) 30 TV TV TV TV TV TV Moderate (60-119) 90 x 2.7 x 2.5 x 2.5 x 4.0 x 2.5 x 2.5 Hard (120-179) 150 x 4.2 x 3.7 x 3.9 x 7.6 x 3.9 x 3.9 Very Hard (180-240) 210 x 5.7 x 4.9 x 5.2 x 11.8 x 5.2 x 5.2 Extremely Hard (400) 400 x 10.0 x 8.4 x 9.0 x 26.7 x 9.0 x 9.0

Table A2-5. General hardness algorithms describing guideline values, where: HMTV = hardness modified trigger value; TV = trigger value; H = measured hardness of the water body.

Metal Hardness-dependent algorithm 0.89 Cd HMTV = TV (H/30) 0.82 Cr (III) HMTV = TV (H/30) 0.85 Cu HMTV = TV (H/30) 1.27 Pb HMTV = TV (H/30) 0.85 Ni HMTV = TV (H/30) 0.85 Zn HMTV = TV (H/30)

______Wetland Research & Management ______44 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Appendix 3. PCA results for physico-chemical variables

(a) PCA on phys-chem for all Ludlow/Tiger Gully sites and all data

Eigen values

PC Eigenvalues %Variation Cum.%Variation 1 16.12 52.0 52.0 2 7.11 22.9 74.9 3 5.35 17.3 92.2 4 1.80 5.8 98.0 5 0.39 1.3 99.2

Eigen vectors (Coefficients in the linear combinations of variables making up PC's)

Variable PC1 PC2 PC3 PC4 PC5 DO% -0.117 -0.010 0.259 -0.465 -0.114 DOmg -0.139 0.048 0.148 -0.540 -0.226 Temp 0.070 -0.062 0.349 -0.046 0.780 Econd 0.228 0.111 -0.080 -0.142 0.032 TDS 0.228 0.109 -0.081 -0.144 0.033 pH -0.081 -0.254 0.262 -0.116 0.212 Al -0.178 0.250 -0.088 0.014 0.059 Alk 0.182 -0.227 -0.123 -0.100 0.017 Ba 0.142 -0.182 -0.273 -0.151 0.075 Ca 0.224 0.161 -0.028 -0.026 0.025 Cd -0.178 0.250 -0.088 0.014 0.059 Cl 0.239 0.089 -0.061 -0.049 0.028 Fe -0.020 -0.334 -0.184 -0.108 0.029 HCO3 0.182 -0.227 -0.123 -0.100 0.017 Hard 0.231 0.140 0.010 -0.013 0.009 K 0.241 -0.061 0.081 -0.006 -0.038 Mg 0.233 0.125 0.042 -0.002 -0.004 Mn -0.100 0.308 -0.172 -0.025 0.094 NH3 -0.149 -0.075 -0.326 -0.123 0.108 NO3 -0.237 -0.067 -0.107 -0.019 0.033 TN 0.073 -0.279 0.255 0.066 -0.120 Na 0.226 0.069 0.160 0.042 -0.051 Ni 0.186 0.239 -0.077 -0.032 0.050 TP 0.011 -0.299 -0.248 -0.135 0.055 SO4 0.221 0.170 -0.021 -0.021 0.023 TSS 0.164 -0.187 -0.230 -0.137 0.059 Zn 0.099 0.107 -0.365 -0.154 0.138 ChnW -0.228 0.036 -0.102 0.046 0.349 WetW -0.215 -0.010 -0.170 0.157 0.144 AveD -0.239 -0.009 -0.073 -0.077 -0.181 MaxD -0.095 0.220 0.061 -0.509 0.148

Principal Component Scores

Sample SCORE1 SCORE2 SCORE3 SCORE4 SCORE5 1 8.305 3.471 -0.824 -0.150 0.039 2 -4.991 3.100 0.290 -2.331 0.236 3 -4.307 2.660 -1.814 2.413 -0.161 4 -0.103 -0.822 3.485 0.692 0.289 5 0.473 -0.709 3.664 0.257 -0.481 6 0.007 -2.461 -1.898 -0.555 -1.060 7 0.237 -2.458 -1.513 -0.598 0.036 8 0.378 -2.781 -1.388 0.273 1.103

______Wetland Research & Management ______45 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

(b) PCA on phys-chem for sites 2 – 8 only and all data

Eigen values

PC Eigenvalues %Variation Cum.%Variation 1 18.92 61.0 61.0 2 8.40 27.1 88.1 3 2.68 8.6 96.8 4 0.52 1.7 98.4 5 0.33 1.1 99.5

Eigenvectors (Coefficients in the linear combinations of variables making up PC's)

Variable PC1 PC2 PC3 PC4 PC5 DO% -0.001 0.196 -0.489 0.056 0.269 DOmg -0.067 0.125 -0.527 0.183 0.263 Temp 0.129 0.216 -0.081 -0.726 -0.065 Econd 0.162 -0.182 -0.279 0.012 -0.071 TDS 0.161 -0.184 -0.278 0.007 -0.064 pH 0.180 0.143 -0.188 -0.340 -0.156 Al -0.227 0.048 -0.007 -0.014 -0.093 Alk 0.182 -0.210 -0.032 -0.023 0.061 Ba 0.106 -0.304 -0.059 -0.049 0.021 Ca 0.223 0.076 0.033 0.038 0.101 Cd -0.227 0.048 -0.007 -0.014 -0.093 Cl 0.211 -0.136 -0.014 -0.005 0.079 Fe 0.122 -0.291 -0.054 -0.045 0.029 HCO3 0.182 -0.210 -0.032 -0.023 0.061 Hard 0.213 0.125 0.043 0.047 0.101 K 0.229 0.013 0.020 0.026 0.099 Mg 0.205 0.150 0.049 0.051 0.099 Mn -0.229 -0.004 -0.018 -0.025 -0.098 NH3 -0.120 -0.291 -0.073 -0.070 -0.074 NO3 -0.207 -0.145 -0.047 -0.050 -0.100 TN 0.222 0.083 0.035 0.040 0.101 Na 0.187 0.196 0.057 0.058 0.095 Ni -0.227 0.048 -0.007 -0.014 -0.093 TP 0.098 -0.310 -0.060 -0.051 0.017 SO4 0.206 0.149 0.048 0.051 0.100 TSS 0.132 -0.281 -0.052 -0.042 0.034 Zn -0.093 -0.312 -0.075 -0.072 -0.064 ChnW -0.206 -0.046 0.037 -0.428 0.483 WetW -0.187 -0.128 0.179 -0.187 0.534 AveD -0.209 -0.049 -0.122 0.231 0.346 MaxD -0.137 0.106 -0.444 -0.101 -0.204

Principal Component Scores

Sample SCORE1 SCORE2 SCORE3 SCORE4 SCORE5 2 -6.089 1.251 -2.761 -0.234 -0.017 3 -6.499 -0.064 2.709 0.212 -0.072 4 2.949 3.547 0.603 -0.448 0.930 5 3.674 3.606 -0.032 0.554 -0.859 6 1.680 -2.909 -0.383 1.092 0.348 7 2.079 -2.825 -0.613 -0.042 0.138 8 2.207 -2.605 0.476 -1.135 -0.468

______Wetland Research & Management ______46 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

(c) PCA results for physico-chemical variables measured in situ in the Ludlow River and Tiger Gully

Eigen values

PC Eigenvalues %Variation Cum.%Variation 1 5.21 52.1 52.1 2 2.54 25.4 77.5 3 1.67 16.7 94.2 4 0.38 3.8 98.0 5 0.11 1.1 99.1

Eigen vectors (Coefficients in the linear combinations of variables making up PC's)

Variable PC1 PC2 PC3 PC4 PC5 DO% -0.265 0.470 -0.135 0.156 -0.447 DOmg -0.296 0.357 -0.336 0.247 -0.261 Temp 0.083 0.491 0.325 -0.680 -0.051 Econd 0.399 0.046 -0.295 -0.047 -0.278 TDS 0.398 0.047 -0.294 -0.043 -0.288 pH -0.193 0.375 0.482 0.233 0.112 ChnW -0.384 -0.225 -0.065 -0.477 -0.279 WetW -0.352 -0.354 0.017 -0.231 -0.318 AveD -0.418 -0.148 -0.090 0.161 0.125 MaxD -0.185 0.259 -0.587 -0.296 0.602

Principal Component Scores

Sample SCORE1 SCORE2 SCORE3 SCORE4 SCORE5 1 4.882 -0.032 -1.439 -0.264 -0.001 2 -3.140 1.354 -2.270 -0.281 0.082 3 -1.439 -3.043 0.036 -0.368 0.184 4 -0.622 1.348 1.321 -0.490 -0.423 5 0.352 1.956 0.951 0.450 0.471 6 -0.346 -0.883 -0.106 1.266 -0.071 7 0.016 -0.289 0.163 0.195 -0.514 8 0.298 -0.410 1.344 -0.509 0.273

______Wetland Research & Management ______47 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

(d) PCA results for physico-chemical variables measured in situ in the Ludlow River, Tiger Gully and Capel River

PCA Ordination on physico-chemical variables measured in situ (a) in the Ludlow/Tiger Gully system in November 2005 and (b) including Capel River data collected in January 2006.

(a)

4

3

2 5 2 4 1

2 C

P 0 1 78 -1 6

-2

-3 3

-4 -8 -6 -4 -2 0 2 4 6 8 10 PC1

(b) 4

3 9 2 3

1 2 2

C 6 P 0 7 8 -1 1 4 -2 5 -3

-4 -8 -6 -4 -2 0 2 4 6 8 10

PC1

Code: p control sitesr controls for Cloverdale but exposed to Yoganup & Yoganup West p exposed sites for Cloverdale, Yoganup West and Yoganup; ø Capel River (Nov. 06).

______Wetland Research & Management ______48 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Appendix 4. Macroinvertebrate taxa occurrences across sites.

Abundances are log10 scale classes: 1 = 1 individual, 2 = 1- 10, 3 = 11 – 100, 4 = 101 - 1000, 5 = >1000. Cons. Cat. = conservation category where: C = common taxa recorded from other states/territories; I = indeterminate; I* = indeterminate but probably occurs only in SW; S = endemic to SW but commonly occurring; L = endemic to SW but with a restricted distribution; Exot = exotic/introduced; VU = listed as vulnerable due to habitat loss and fragmentation of populations (CALM Priority 4; IUCN Redlist 2004). ÖTaxonomy of family and/or genera under review (Brenton Knott, UWA, pers. comm.). Cons SITES TAXA Cat 1 2 3 4 5 6 7 8 9

TURBELLARIA Turbellaria sp. C ------1 -- -- TEMNOCEPHALA Temnocephalida sp. C ------1 ------NEMATODA Nematoda spp. C -- 3 ------4 2 ANNELIDA ------OLIGOCHAETA Oligochaeta spp. C 2 3 2 3 -- 2 3 3 3 HIRUDINEA Glossophonidae spp. C -- 2 ------3 -- -- Hirudinidae sp. C -- -- 1 ------MOLLUSCA Hyriidae Westralunio carteri L, VU ------2 Ancyclidae Ferrissia (pettancylus) petterdi C -- -- 1 2 3 3 4 2 Planorbidae Glyptophysa (Glyptophysa) sp. I -- 3 -- 2 2 1 2 1 -- ARTHROPODA ARACHNIDA ACARIFORMES ACARINA Unidentified larvae/nymphs I -- -- 1 2 1 -- 2 -- -- ORIBATIDA Oribatida spp. I 2 ------1 -- -- PROSTIGMATA Halacaridae Halacaridae sp. I ------1 ------Arrenuridae Arrenurus sp. C ------2 -- -- Eylaidae Eylais sp. I ------2 -- -- Hydrachnidae Hydrachna sp. (nymph) I ------1 -- -- Hygrobatidae Procorticacarus sp. C ------2 -- -- Australiobates sp. ------2 Limnesiidae Limnesia longigenitalis C ------2 -- -- Pionidae Pionidae sp. (nymph) S ------1 -- Trombidioidea Trombidioidea sp. I ------1 ------PARASITIFORMES ------MESOSTIGMATA Mesostigmata spp. (terrestrial?) I ------2 -- -- CRUSTACEA DECAPODA Parastacidae Cherax plebejus S ------2 3 -- -- 3 Cherax quincecarinatus S -- -- 3 ------2 3 Cherax sp. S 2 ------2 -- -- Palaeomonetes australisÖ C ------2 CLADOCERA Cladocera spp. I 1 3 -- 4 -- -- 3 3 2 COPEPODA Cyclopoida spp. I 4 4 3 4 3 3 5 4 4 OSTRACODA Ostracoda spp. I 2 5 2 3 2 2 4 3 2 AMPHIPODA Perthidae Austrochiltonia subtenuis S ------1 -- -- Perthia branchialisÖ S ------2 3 -- Perthia acutitelsonÖ S ------1 -- -- Perthia sp.Ö S ------2 COLLEMBOLLA Entomobryoidea spp. C 1 -- -- 3 -- -- 1 -- -- Poduroidea spp. C ------2 INSECTA DIPTERA Ceratopogonidae Ceratopogoniinae sp. I 1 ------Forcypomiinae sp. I ------1 -- 3 Chironomidae Chironomidae spp. (pupa) I ------2 2 2 3 -- 1 Chironominae Chironomus aff. alternans C -- 3 3 -- 2 1 3 3 3 Cryptochironomus griseidorsum C 2 -- -- 1 2 -- 2 -- -- Dicrotendipes sp. V47 S -- -- 1 ------Dicrotendipes conjunctus C 3 -- -- 3 ------

______Wetland Research & Management ______49 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Cons SITES TAXA Cat 1 2 3 4 5 6 7 8 9 Harrisius sp. V27 I* ------1 Harrisius sp. V40 I* -- -- 2 ------1 Parachironomous sp. VSCL35 I ------1 Paracladopelma sp. VCD10 I -- -- 1 -- -- 2 2 -- 5 Polypedilum nubifer C ------3 1 1 ------Polypedilum watsoni C ------1 Polypedilum sp. V3 I -- 1 1 ------2 1 3 Riethia sp. V4 I ------3 Riethia sp. V5 I ------2 Rheotanytarsus sp. I -- -- 2 2 -- 3 3 3 -- Rheotanytarsus sp. V18 I ------3 Tanytarsus sp. 1 I 2 2 2 1 2 3 3 3 -- Tanytarsus sp. 2 I 4 1 -- 2 2 3 3 2 -- Tanytarsus sp. I ------3 Orthocladiinae Botryocladius bibulmun S ------2 Corynoneura sp. C 3 1 -- 1 -- 2 2 -- -- Cricotopus albitarsis C ------2 ------Cricotopus annuliventris C 3 -- 2 3 3 3 3 3 2 Limnophyes pullulus C 2 1 1 -- -- 2 ------Thienemanniella sp. C 3 1 2 3 3 4 3 3 3 Tanypodinae Paramerina levidensis S 2 2 3 -- 2 3 3 2 3 Procladius paludicola C ------2 1 ------1 Procladius villosimanus C -- 2 -- -- 2 2 ------Culicidae Anopheles sp. C 2 -- -- 1 1 1 2 1 1 Culicidae spp. I ------2 -- -- Culicidae sp. (pupa) I 2 1 2 1 2 2 2 -- -- Empididae Empididae spp. I ------1 2 2 -- Muscidae Muscidae sp. (pupa) I 1 ------Psychodidae Psychodidae sp. I ------1 -- 2 Simulidae Austrosimulium sp. C -- -- 2 ------1 -- -- Simulium ornatipes Exot 3 -- -- 4 4 3 3 5 2 Simulium ornatipes (pupa) Exot ------3 1 -- -- 2 1 Tabanidae Tabanidae sp. I -- 1 ------Tipulidae Tipulidae spp. I 2 1 2 -- -- 1 2 2 3 Athericidae sp. I ------1 ODONATA ZYGOPTERA Zygoptera sp. (juvenile) I ------1 -- 2 ANISOPTERA Telephlebiidae Austroaeschna anacantha S ------2 Hemicorduliidae Hemicordulia sp. (immature) C ------1 TRICHOPTERA Hydrobiosidae Taschorema pallescens S ------2 Hydropsychidae Smicrophylax australis S ------3 Hydroptilidae Acritoptila/Hellyethira spp. I* 3 3 -- 3 -- 3 3 2 2 Oxyethira sp. I* 2 ------Hydroptilidae spp. (juvenile) I 3 ------2 -- Leptoceridae Oecetis sp. I* -- 2 -- 2 2 2 2 2 2 Notalina spira C 1 ------Triplectides australis C 2 2 -- 2 2 2 2 2 -- Leptoceridae spp. (juvenile) I -- 2 -- 2 1 -- 2 -- 2 EPHEMEROPTERA Baetidae Cloeon sp. C -- -- 1 ------Caenidae Tasmanocoensis tillyardi C ------2 Leptophlebiidae ?Neboissophlebia sp. I* ------1 Leptophlebiidae spp. I* ------3 HEMIPTERA Corixidae Agraptocorixa parvipunctata C ------2 -- Agraptocorixa sp. (female) C ------2 1 ------Micronecta robusta C 1 ------1 -- -- Micronecta sp. (juvenile) C ------2 -- -- Sigara truncatipala C -- 1 -- 2 -- 2 2 1 --

______Wetland Research & Management ______50 Cloverdale Project Baseline Aquatic Biology & Water Quality Study 2005/06

Cons SITES TAXA Cat 1 2 3 4 5 6 7 8 9 Sigara sp. (female) C 2 2 -- 1 -- 2 2 -- -- Corixidae spp. (juvenile) C 2 3 -- 2 2 2 4 3 -- Mesoveliidae Mesoveliidae sp. I ------1 -- -- Notonectidae Anisops thienemanni C -- 2 -- 2 -- -- 2 2 -- Anisops sp. 1 ?C ------2 ------Anisops sp. (females & juveniles) C 2 2 -- 3 2 2 3 3 -- Veliidae Microvelia sp. I 2 ------1 Veliidae spp. (juvenile) I 2 -- -- 1 ------2 COLEOPTERA Dytiscidae Allodessus bistrigatus C 1 ------Antiporus femoralis C 1 ------1 1 -- Antiporus gilberti C ------1 ------1 -- Antiporus sp. (larva) I ------1 -- -- 1 -- -- Lancetes lanceolatus C -- 2 ------2 ------Limbodessus inornatus S 3 2 -- 1 1 -- 2 2 2 Limbodessus shuckhardi C ------1 Megaporus howitti C -- 1 -- 1 -- 1 1 -- -- Megaporus solidus C ------1 Megaporus sp. (larva) I ------1 1 -- Necterosoma darwini S 2 1 -- 1 1 2 -- 2 2 Necterosoma sp. (larva) I 3 -- 2 2 2 2 3 3 -- Paroster sp. (female) I 1 ------1 -- -- Platynectes aenescens C 1 ------Platynectes decempunctalis (var. C 2 1 -- -- 1 2 1 -- -- polygrammus) Platynectes sp. (larva) I 3 1 2 3 2 3 2 2 -- Rhantus suturalis C 3 3 2 1 2 3 2 2 -- Rhantus sp. (larva) I -- 2 -- 1 -- -- 1 2 -- Sternopriscus brownii S 1 1 ------Sternopriscus multimaculatus C 2 ------Sternopriscus sp. (larva) I 2 2 -- -- 2 ------1 Tribe Bidessini spp. (larva) I ------1 -- -- Gyrinidae Macrogyrus sp. I ------1 Hydraenidae Octhebius sp. I ------1 Hydrochidae Hydrochus sp. I ------1 Hydrophilidae Enochrus eyrensis C 1 ------Limnoxenus zealandicus C 2 1 -- 1 ------Limnoxenus sp. (larva) C 2 1 -- 2 2 1 2 2 -- Paracymus pygmaeus C 3 1 -- 1 1 -- 2 -- 2 Paracymus ?spenceri C 2 ------Tribe Anacaenini sp. (larva) I ------1 ------Hydrophilidae spp. (larva) I 2 ------Total species 52 41 25 49 35 42 69 42 59

______Wetland Research & Management ______51