sign in sign up
<<
Home , Leycester Creek, Orara River, Richmond River, River ecosystem, Shannon Brook

AN INVESTIGATION OF THE POTENTIAL ECOLOGICAL IMPACTS OF FRESHWATER EXTRACTION FROM THE RICHMOND TIDAL POOL

WRL Technical Report 99/51 November 1999 THE QUALITY OF THIS SCAN IS BASED ON THE ORIGNAL ITEM THE UNIVERSITY OF WATER RESEARCH LABORATORY

AN INVESTIGATION OF THE POTENTIAL ECOLOGICAL IMPACTS OF FRESHWATER EXTRACTION FROM THE TIDAL POOL

WRL Technical Report 99/51 November 1999

by

W L Peirson, K A Bishop, R Nittim and M J Chadwick

https://doi.org/ 10.4225/53/58d49e7613006 Unisearch Ltd a c n ooo 263 025 Water Research Laboratory Technical Report No 99/51 University of New South Wales Report Status Final King Street Date of Issue November 1999 Manly Vale NSW 2093

Telephone: +61 (2) 9949 4488 WRL Project No. 99069 Facsimile: +61 (2) 9949 4188 Project Manager W L Peirson

Title An Investigation of the Potential Ecological Impacts of Freshwater Extraction from the Richmond River Tidal Pool

Author(s) W L Peirson, K A Bishop, R Nittim and M J Chadwick

Client Name Department of Land and Water Conservation, NSW.

Client Address AMP Centre, 24 Gordon St, PO Box 582, NSW 2450.

Client Contact Linden Bird

Client Reference

The work reported herein was carried out by Unisearch Ltd, the commercial company of the University of New South Wales, acting on behalf of the client. The work was undertaken at the University's Water Research Laboratory.

Information published in this report is available for general release only by permission of the Water Research Laboratory and the client. WRL TECHNICAL REPORT 99/51

CONTENTS

1. INTRODUCTION 1 2. HYDROLOGICAL ANALYSIS 3 2.1 Approach 3 2.2 Description 3 2.3 Hydrological Model Calibration 4 2.4 Hydrological Model Results 6 2.5 Model Description 7 2.6 Estuary Model Configuration 7 2.7 Estuary Model Calibration 8 2.7.1 RMA-2 Calibration 8 2.7.2 RMA-2 Verification 10 2.7.3 RMA-11 Calibration 10 2.8 Prediction of Salinity Structure 12 2.8.1 Empirical Calculations 12 2.8.2 Model Simulations and Management Scenarios 12 2.8.3 Discussion of Results 13 2.9 Simulation of 1901-1902 15 3. ANALYSIS OF 17 3.1 Approach 17 3.2 Estuarine Vegetation 17 3.3 Communities and Their 19 3.4 The Platypus 21 3.5 Indicative Salinity Limits and Their Use 22 3.6 Risk Analysis of Water Extraction Scenarios 23 3.6.1 Dimensions of the Analysis 23 3.6.2 Habitat-Availability Data Arising from Salinity Thresholds 25 3.6.3 Assumptions Regarding Risk 26 3.6.4 Measuring Habitat Contraction and Risk Categories 26 3.6.5 Percentile Condition Synthesis and Thresholds 27 3.6.6 Changes in Habitat Availability and Risk 28 3.6.7 Synthesis 34 4. CONCLUSIONS AND RECOMMENDATIONS 38 5. REFERENCES 41

LIST OF TABLES

2.1 Catchments Selected for AWBM Calibration 2.2 AWBM Catchment Models: Key Parameters Adopted for Calibration 2.3 Comparison of Measured and Modelled Tidal Prisms: Richmond River, 3 November, 1994 2.4 Comparison of Measured and Modelled Tidal Prisms: Wilsons River, 3 November, 1994

- i - WRL TECHNICAL REPORT 99/51

LIST OF FIGURES

1.1 General Plan of the Richmond Catchment 2.1 Schematic Representation of the AWBM Model 2.2 Catchment Plan Showing Major Rainfall and Gauge Locations 2.3 AWBM Model Calibration: Richmond River at Casino 2.4 AWBM Model Calibration: Leycester River at Rock 2.5 AWBM Model Calibration: Wilsons River at Eltham 2.6 AWBM Model Calibration: Myrtle Creek at Rappville 2.7 AWBM Model Verification: Richmond River at Casino 2.8 AWBM Model Verification: Leycester River at Rock Valley 2.9 AWBM Model Verification: Wilsons River at Eltham 2.10 AWBM Model Verification: Myrtle Creek at Rappville 2.11 Representative Rainfall and Streamflow (1940-1997) 2.12 Volume and Area Richmond River 2.13 Channel Volume and Area Wilsons River 2.14 Channel Volume and Area Bungawalbin Creek 2.15 The Finite Element Estuary Mesh 2.16 Locations of Calibration Data for RMA-2 Calibration: 3 November 1994 2.17 Comparison of Measured and Modelled Tide Levels - Richmond River: 3 Nov 1994 2.18 Comparison of Measured and Modelled Tide Levels - Wilsons River: 3 Nov 1994 2.19 Comparison of Measured and Modelled Tidal Lags - Richmond River: 3 No. 1994 2.20 Comparison of Measured and Modelled Tidal Lags - Wilsons River: 3 Nov 1994 2.21 Comparison of Measured and Modelled Salt Distributions: Richmond River, 1994 2.22 Existing Conditions: Modelled Salt Distributions in the Richmond River: 1940-1997 2.23 Existing Conditions: Modelled Salt Distributions in the Wilsons River: 1940-1997 2.24 Existing Conditions: Modelled Salt Distributions in Bungawalbin Creek: 1940-1997 2.25 Option A: Modelled Salt Distributions in the Richmond River: 1940-1997 2.26 Option A: Modelled Salt Distributions in the Wilsons River: 1940-1997 2.27 Option A: Modelled Salt Distributions in Bungawalbin Creek: 1940-1997 2.28 Option B: Modelled Salt Distributions in the Richmond River: 1940-1997 2.29 Option B: Modelled Salt Distributions in the Wilsons River: 1940-1997 2.30 Option B: Modelled Salt Distributions in Bungawalbin Creek: 1940-1997 2.31 Option C: Modelled Salt Distributions in the Richmond River: 1940-1997 2.32 Option C: Modelled Salt Distributions in the Wilsons River: 1940-1997 2.33 Option C: Modelled Salt Distributions in Bungawalbin Creek: 1940-1997 2.34 Option D: Modelled Salt Distributions in the Richmond River: 1940-1997

- n - WRL TECHNICAL REPORT 99/51

List of Figures (Contd)

2.35 Option D: Modelled Salt Distributions in the Wilsons River: 1940-1997 2.36 Option D: Modelled Salt Distributions in Bungawalbin Creek: 1940-1997 2.37 Option E: Modelled Salt Distributions in the Richmond River: 1940-1997 2.38 Option E: Modelled Salt Distributions in the Wilsons River: 1940-1997 2.39 Option E: Modelled Salt Distributions in Bungawalbin Creek: 1940-1997 2.40 Option F: Modelled Salt Distributions in the Richmond River: 1940-1997 2.41 Option F: Modelled Salt Distributions in the Wilsons River: 1940-1997 2.42 Option F: Modelled Salt Distributions in Bungawalbin Creek: 1940-1997 2.43 Modelled Salt Distributions in the Richmond River, 1902 3.1 Changes in Salinity Percentiles Along the Main Arm of the Richmond River Estuary: Existing Condition 3.2A Length of Habitat Remaining for Oxleyan Pygmy Perch (Thresh. =0.12 ppt) under Existing Conditions and 6 Options : Richmond River (middle) Arm 3.2B Length of Habitat Remaining for Oxleyan Pygmy Perch (Thresh. = 0.12 ppt) under Existing Conditions and 6 Options : Wilsons River Arm 3.2C Length of Habitat Remaining for Oxleyan Pygmy Perch (Thresh. = 0.12 ppt) under Existing Conditions and 6 Options : Bungawalbin Creek Arm 3.3A Length of Habitat Remaining for the Platypus (Thresh. = 0.5 ppt) under Existing Conditions and 6 Options : Richmond River (middle) Arm 3.3B Length of Habitat Remaining for the Platypus (Thresh. = 0.5 ppt) under Existing Conditions and 6 Options : Wilsons River Arm 3.3C Length of Habitat Remaining for the Platypus (Thresh. = 0.5 ppt) under Existing Conditions and 6 Options : Bungawalbin Creek Arm 3.4A Area of FE+ETC+Macrophytes Remaining (Thresh. = lppt) under Existing Conditions and 6 Options : Richmond River (middle) Arm 3.4B Area of FE+ETC+Macrophytes Remaining (Thresh. = lppt) under Existing Conditions and 6 Options : Wilsons River Arm 3.4C Area of FE+ETC+Macrophytes Remaining (Thresh. = lppt) under Existing Conditions and 6 Options : Bungawalbin Creek Arm 3.5 A Area of Habitat Remaining for Adult Bass NS & RG (Thresh. = 5 ppt) under Existing Conditions and 6 Options : Richmond River (middle) Arm 3.5B Area of Habitat Remaining for Adult Bass NS & RG (Thresh. = 5 ppt) under Existing Conditions and 6 Options : Bungawalbin Creek Arm 3.6 Area of Habitat Remaining for Bass Spawning (Thresh. = BT. 8-13 ppt) under Existing Conditions and 6 Options : Richmond River (middle) Arm 3.7 Length of Habitat Remaining for SR Oysters with LMF (Thresh. = <20 ppt) under Existing Conditions and 6 Options : Richmond River (middle) Arm 3.8 Summary of Percentage Change in the Length/Area of Estuary with Salinites Suitable for Six Facets 3.9 Relationships between the Risk Index and the Extraction Rate Surrogate (Ml/Ha/yr) for Four Freshwater-Content Percentile Conditions of the Richmond River Estuary

- in - WRL TECHNICAL REPORT 99/51 1 .

1. INTRODUCTION

This is a detailed assessment of the impacts of extracting from below the tidal limit on the saline structure of the Richmond River estuary. This investigation has been undertaken for the North Coast office of the Department of Land and Water Conservation by the Water Research Laboratory in conjunction with a team of aquatic ecologists.

This study is a component of a wider program currently being undertaken by the New South Wales Government. Through the Council of Australian Governments, it has been agreed that action is needed Australia-wide to minimise unsustainable use of water resources. As a consequence, the State government is setting interim environmental objectives for water quality and quantity which all planning processes will aim to achieve. The 'stress' of a given stream is assessed in terms of these objectives so that independent and transparent management decisions are made.

In , the use of water is quantified simply as the proportion diverted for water supply, or industrial uses of the total flow down the river.

This focus of this investigation is the Richmond River estuarine system in northern New South Wales (Figure 1.1). The issue that is to be determined from this investigation is whether irrigators below the tidal limits of should be treated differently from their counterparts extracting water from the rivers upstream.

During periods of low freshwater inflow to an estuary, saline waters enter from the ocean through the estuary mouth. These saline waters enter as density currents or as a result of tidal mixing. During periods of high freshwater inflow from the catchments salt water is flushed from the estuary. The hydrological nature of the Australian climate means that significantly different saline structures can be observed in any given estuary depending on the antecedent rainfall.

Downstream of its tidal limits, large pools of freshwater can remain in the upstream reaches of an estuary during long periods of low freshwater inflow. These have been termed "tidal pools".

The fundamental objective of this investigation is to assess the impact of freshwater extractions from the tidal pool on the estuarine behaviour and aquatic of the Richmond River system. In particular, this investigation is to: WRL TECHNICAL REPORT 99/51 2.

• determine how the saline structure of the Richmond River would change as a consequence of changing the extraction rates of water for irrigation;

• identify the estuarine aquatic ecology likely to be affected by changes in saline structure;

• assess the risks to the aquatic ecology as a result of changing extraction rates from the tidal pool.

Three tasks have been undertaken:

1. Development and application of a numerical model of the estuary capable of reproducing the changes in longitudinal distribution of salinity in response to changing freshwater inflows and climatic conditions (this is discussed in Section 2);

2. A desktop identification of the salinity-exposure characteristics (salinity concentration and duration) which best typify transition conditions where freshwater biota would be displaced by more salt-tolerant estuarine biota in the Richmond River estuary (this is reported in Section 3); and,

3. Discussion of the findings of the first two tasks with recommendations of a management strategy for tidal pool extractions. WRL TECHNICAL REPORT 99/51 3.

2. HYDROLOGICAL ANALYSIS

2.1 Approach

To assess the long-term distributions of salinity in the Richmond River, a numerical model of the estuary has been developed capable of reproducing the changes in longitudinal distribution of salinity in response to changing freshwater inflows and climatic conditions.

This estuary model has been calibrated against available data and used to predict salinity distributions in the Richmond River estuary from January 1940 to December 1997.

The salinity distributions are strongly dependent on the freshwater inflows to the estuary. Consequently, it is essential that the runoff characteristics of the Richmond catchment be carefully assessed - this has been undertaken using a hydrological model of the catchments calibrated against available streamflow records.

Using the hydrological model, these freshwater inflows have been derived for 1994, a period when a suite of salinity observations was recorded on the Richmond. The estuary model has been calibrated against these observations.

The establishment, calibration, verification and results of the hydrological model and the estuary model are described in this chapter.

2.2 Hydrological Model Description

The freshwater flows in New South Wales rivers do not follow a consistent annual pattern. Consequently,'for reliable assessments of changing estuarine saline structure, historical analyses must be undertaken.

Therefore, a means of determining reliable estimates of freshwater input to the estuary for a long historical period had to be determined. In this study, this has been undertaken for the period extending from 1 January 1940 to 31 December 1997.

Daily streamflow records in the Richmond Valley extend back to:

• 1943 on Richmond River at Casino; • 1951 on Leycester River at Rock Valley; • 1957 on Wilsons River at Eltham; and, • 1969 on Myrtle Creek at Rappville.

The locations of these stations are indicated in Figure 1.1. WRL TECHNICAL REPORT 99/51 4.

It is to be noted that these records cover the -dominated era which started about 1950 but not the drought-dominated era from 1900 to 1950.

Daily rainfall records extend back to:

• 1858 at Casino; • 1884 at Cumbalum near the coast and Lismore further inland; • 1903 at Nimbin at the base of the Border Ranges; • 1910 at Old Benalbo west of the Richmond Range and outside the catchment; • 1936 at Bingeebeebra; and • 1940 at Mt Pickapene both on the eastern side of the Dividing Range.

Accordingly a good daily rainfall record is available from 1940 onwards. This rainfall record together with the daily flow records has been used to calibrate hydrological models for the various catchments. The calibrated model has been used to extend the flow records for all catchments back to 1940.

The AWBM model (Boughton 1993, Boughton and Carrol 1993) has been used to generate the daily flows. This model was selected because it was developed in Australia and has been calibrated on a large number of catchments in Eastern Australia.

The AWBM model is a saturated overland flow model which allows for variable source areas of in different storms and in different periods of a given storm. The component simulates the recharge and of shallow stores.

The operation of the AWBM model is illustrated in Figure 2.1. In this application it has been used as an eight parameter model so that the baseflow component resulting from antecedent rainfall can be determined.

The AWBM model takes precipitation and evaporation rates as inputs and determines surface runoff as the excess of the baseflow recharge. Baseflow discharge is determined in terms of a daily recession constant.

2.3 Hydrological Model Calibration

Initially, the AWBM model was established for four representative catchments within the study area, each of which had suitable streamflow data available. The catchments used for calibration are listed in Table 2.1 and are shown in Figure 2.2. Using 1:100000 topographic WRL TECHNICAL REPORT 99/51 5. maps of the study area, catchment boundaries were delineated and the corresponding catchment areas determined (Table 2.1).

Table 2.1 Catchments Selected for AWBM Calibration

Catchment Area (km2) Gauging Station Casino 1872 203004 Richmond River at Casino Eltham 220 203014 Wilsons River at Eltham 178 203010 Leycester Creek at Rock Valley Myrtle Creek 355 203030 Myrtle Creek at Rappville

Using daily rainfall records from the relevant rainfall stations, a mean daily rainfall depth was determined using the Theissen method. Mean monthly evaporation data supplied by the Bureau of Meteorology for the Alstonville Tropical Fruit Research Station (Station No. 058131), was incorporated into each of the catchment models. Daily streamflow data for each of the gauging stations listed in Table 2.1 was obtained from the DLWC data archive, Pinneena v6.0.

Using this data, an AWBM model for each calibration catchment was created. Each model was then calibrated over the period 1990-1991 to determine optimum values for the capacities and distribution of surface stores, daily surface recession constant and baseflow parameters (baseflow index and daily recession constant). The calibration period (1990-91) was chosen as it included two and a prolonged drought. It was found that it was possible to tune the model to obtain good representation of either the larger flow events or period of low freshwater flow but not both. Given that the periods of primary interest to this investigation are sustained dry periods, it was decided to focus model calibration on periods of low freshwater flow. Table 2.2 gives the parameters adopted for calibration for the catchments in Table 2.1.

Table 2.2 AWBM Catchment Models: Key Parameters Adopted for Calibration

Catchment Area Surface Stores Ks Baseflow Parameters

(km2) Capacities (mm) Partial Areas (d-1) BFI Kr (d ')

Casino 1872 5, 40, 120 0.049, 0.514, 0.437 0.3 0.416 0.970 Eltham 220 5, 140, 200 0.342, 0.512, 0.146 0.3 0.591 0.970 Leycester Ck 178 5, 160, 220 0.202, 0.751,0.047 0.3 0.312 0.962 Myrtle Ck 355 3, 120, 140 0.010, 0.840, 0.150 0.5 0.300 0.958 Ks: Daily Surface Flow Recession Constant BFI: Baseflow Index KR: Daily Baseflow Recession Constant WRL TECHNICAL REPORT 99/51 6.

Model calibration results for 1991-1992 are shown in Figures 2.3 to 2.6. The hydrograph recession rates are not always identical and the model has been tuned to reproduce an average behaviour during periods of low rainfall. It is to be noted that the much larger catchment of the Richmond River at Casino means that this provides the dominant freshwater inflow.

Figures 2.7 to 2.10 show the calibrated AWBM model results for each of the four catchments for 1994. This is of particular significance as several water quality surveys of the Richmond River were conducted in that year. The data from these surveys was then used to calibrate the water quality model discussed in Section 2.7.2. It should be noted that the initial conditions for these runs were arbitrary, hence the relatively poor match between modelled and observed during the initial month(s) of the simulation. However, this feature has no significant bearing on the results produced for the full hydrological simulation of Jan 1940 to Dec 1997.

2.4 Hydrological Model Results

For the Casino catchment, the calibrated model described in Section 2.3 was used to estimate streamflow for the entire simulation period. For the remaining catchments in the study region, two approaches were adopted to generate streamflow hydrographs.

The first approach assumed that certain catchments shared the same physical characteristics as one of the catchments chosen for calibration. This enabled the application of the AWBM parameters derived from the calibration process to larger, ungauged catchments adjacent to the calibration catchments. An increase in the daily baseflow recession constant was made to account for ungauged catchments that were significantly larger in size than the corresponding calibrated catchment.

The remaining ungauged catchments in the study area that could not be defined adequately using the first approach were confined to the lower parts of the catchment. These catchments were dominated by flat and swamp/marsh-like terrain. For these areas, the

baseflow parameters were set as 0.6 and 0.98 for the BFI and K r respectively. Boughton (1993) presented a comparison of average surface storage capacities derived from a previous study of the calibration of 184 rural catchments in south-eastern mainland Australia. Using these results as a basis, the average storage capacity of the remaining catchments was set at 155 mm. This average capacity was then desegregated using the WRL TECHNICAL REPORT 99/51 7.

default partial area definition in AWBM, as per recommendations for modelling ungauged catchments (Boughton, 1996).

A complete set of inflow hydrographs was produced for input to the estuary model. An example hydrograph determined from hydrological modelling for the Casino catchment is shown in Figure 2.11.

2.5 Estuary Model Description

The numerical models to simulate estuary behaviour are two components of the Management Associates (RMA) suite. Both components have used an identical one­ dimensional mesh using appropriate junctions where necessary.

The estuarine flows have been simulated using the model RMA-2. When restricted to one-dimensional elements, this model simulates unsteady ffee-surface flow in prismatic channels. Boundary roughness is represented Manning's n. For more details see King, 1994.

Saline dispersion within the estuary has been simulated in the companion water quality model RMA-11. Saline transport is simulated using the flows computed by RMA-2 with diffusivities used to simulate saline dispersion being specified for RMA-11. RMA-11 is capable of simulating more complex water quality interactions if required.

2.6 Estuary Model Configuration

The model geometry was established from a 1980 hydrographic survey of the river by Public Works. To these, corrections and additional information obtained during the flood study were included.

Representative estuary cross-sections have be determined using the techniques developed for the Tamar estuary by Nittim and Peirson (1987). In this model, each cross-section is represented by the mean width at mid-tide and a mean hydraulic depth. This allows a simple representation of the estuary cross-sections to be adopted whilst preserving the internal volume of the estuary and the phase speed of the tide along the channel. Figures 2.12 to 2.14 show the river channel volumes below 0.0mAHD and channel surface area at 0.0 m AHD of the three major arms of the estuary as they are represented in the RMA model. WRL TECHNICAL REPORT 99/51

The mesh developed for this investigation is shown with annotation in Figure 2.15. Boundary conditions have been applied to this mesh to represent the ocean and freshwater inputs to the model for the major . For those catchment outlets located at the upstream boundaries of the RMA model, freshwater inputs were defined as upstream boundary flow hydrographs. The remaining AWBM catchment outputs were input as element inflows at the appropriate locations along the finite element network, in accordance with the sub-catchment locations in Figure 2.2. The ocean boundary at the downstream end of the model was represented using a constant tailwater boundary condition.

There is some uncertainty regarding the location and magnitude of irrigation extractions from the tidal pool and these have implications for the model calibration and assumed base cases. Based on the data provided by DLWC, it has been assumed that the total annual irrigation rate along Richmond River tidal pool is 2ML/Ha. This rate has been desegregated into varying monthly values by DLWC. For simplicity in this analysis, these monthly rates have been lumped for each tributary and applied at the tidal limit.

2.7 Estuary Model Calibration

2.7.1 RMA-2 Calibration

The hydrodynamic model has been calibrated against the measurements on 3 November 1994. These are described in detail in MHL (1995). The calibration has been refined to replicate recorded tidal levels, discharges and lags through the Richmond River system to the tidal limits.

The location of verification points for the model are shown in Figure 2.16. The model was calibrated by varying the roughness along the tidal channel.

The high and low tide levels along the Richmond and Wilson Rivers are shown in comparison with the recorded data in Figures 2.17 and 2.18. At distances greater than 50km from the ocean, the differences between the model and measured values is less than 0.05 m. In many locations there appears to be a contribution from the adopted survey datum as the offsets for high and low tide are in the same direction. Less than 50 km from the ocean, the error can be as high as 0.1 m.

The corresponding tidal lags for low and high tides are shown in Figures 2.19 and 2.20. If the offset at the mouth is removed, the tidal lags on the main Richmond arm to a distance of 55 km from the ocean (Coraki) are replicated within 15minutes. Beyond Coraki, there is a WRL TECHNICAL REPORT 99/51 9.

jump in the tidal lag which is not replicated in the model and the error in the model estimated tidal lag increases to about 30 minutes. The cause of the jump in the tidal lag at Coraki is unknown.

The measured tidal discharges at key points shown in Figure 2.16 have been replicated in the model. These are summarised in Tables 2.3 and 2.4 below.

Table 2.3: Comparison of Measured and Modelled Tidal Prisms Richmond River, 3 November 1994

Site 5 Site 8 Site 10 Site 13 Site 25 Tidal Prism Model Proto. Model Proto. Model Proto. Model Proto. Model Proto. (m3xl06) Flood 20.66 21.49 2.54 2.31 1.25 1.29 7.09 8.05 1.00 1.93 Ebb 18.49 18.59 2.90 2.15 1.51 1.21 5.96 6.70 1.28 1.88

Table 2.4: Comparison of Measured and Modelled Tidal Prisms Wilsons River, 3 November 1994

Site 19 Site 23 Site 21 Tidal Prism Model Proto. Model Proto. Model Proto. (nTxlO6) Flood 2.46 2.82 0.17 0.51 0.26 0.50 Ebb 2.03 2.33 0.19 0.52 0.23 0.50

The surveyed cross sections were used to estimate the model cross sections and the model results replicate the measured tidal prisms to within 1.0m3xl06. The model does show a tendency to underestimate the tidal prisms near the headwaters of the estuary.

No measurements of the tidal flux at the entrance of Bungawalbin Creek have been made. Only a single measurement of tidal range (0.1m) was available near the tidal limit on Bungawalbin Creek on 22 July 1997. The Manning's n of this tributary was increased to 0.06 to reduce the tidal range in the model at this point to 0.4m. Further increases in the boundary roughness would be unrealistic. The small tidal amplitude observed may be a function of a local and is therefore of limited value. All other aspects of the model representation of Bungawalbin Creek remain uncalibrated.

From the calibration process of the hydrodynamic model, (and apart from the section of Bungawalbin Creek referred to above), a Manning's n value of 0.023 was adopted for all other reaches. This value is consistent with expected values for the degree of channel irregularity. WRL TECHNICAL REPORT 99/51 10.

2.7.2 RMA-2 Verification

Following calibration of the RMA-2 model based on tidal measurements in November 1994, it was originally envisaged that the hydrodynamic model would be verified using tidal discharge measurements collected on 26 May 1977 (MHL, 1978).

However, it became apparent that the data for the 1977 survey was collected several days after a flood event on the Richmond River catchment. Examination of streamflow records from this period showed that the freshwater flows at a number of locations (e.g. Casino, Eltham) were still of some significance at the time of the survey.

An inspection of the tidal data contained in the MHL report reflected the presence of these freshwater inflows - hence it was not possible to adequately define the tailwater levels for model verification under these conditions. In light of these factors, it was not possible to undertake verification of the RMA-2 tidal calibration using the information from the 1977 survey.

2.7.3 RMA-11 Calibration

The computational times required for the long-term simulations require that a model time step of a least a day be used. Consequently, calibration of the water quality model using a short time step that incorporates tidal motions would be inappropriate. RMA-11 calibration has been undertaken in comparison with the salinity structure observations obtained during 1994 on a time step of one day.

As shown by MHL (1998), the estuary does not exhibit strong vertical stratification except during periods of strong freshwater inflow. As periods of low freshwater inflow are of primary interest during this investigation, the model system used has neglected stratification effects. The well-mixed state indicates that boundary friction resulting from tidal flow is the dominant process determining mixing during periods of low flow in the Richmond system.

In order to use a time step of one day, tidal exchange was simulated by using high dispersion coefficients and a fixed high tide level at the river mouth (0.79m AHD).

The calibration of the model has been undertaken using a varying dispersion coefficient to account for the changing geometry and tidal flushing of the system. Following Elder (1959), we have used: WRL TECHNICAL REPORT 99/51 11.

D = aU max,tidal h

where Umax, Mai is the maximum tidal velocity occurring in a given reach, h is the mean hydraulic depth and a is a coefficient.

A single value of a=15 has been used for the entire model with diffusivities ranging from approximately 4.5m2s'1 at the headwaters to 42 m2s-1 at the mouth. This has been calibrated to achieve the best match between the model and the observed longitudinal salinity structure during 1994. The progression of salt along the estuary during 1994 in the model is shown in Figure 2.21. A number of remarks should be made regarding the model behaviour.

First, saline intrusion after fresh events is underestimated by the model. This is a consequence of the stratified behaviour of the estuary (MHL, 1998a). Density currents associated with the more dense ocean waters penetrate along the bed to transport saline waters upstream. Such behaviour can be represented by more sophisticated stratified flow models (for example, King 1996) but is beyond the capability of the simple model used here.

Secondly, once the estuary becomes vertically homogenous the diffusion of salt from the ocean is adequately represented. However, it proved impossible to reproduce the measured spacing of the isopycnals in the model. This may be a consequence of the effects of saline density currents following major fresh events on the estuary.

Higher salinities (>10ppt) are generally restricted to the most downstream reaches of the estuary where irrigation is less likely. Consequently, it was decided to calibrate the model to reproduce the rate of upstream movement of the 0.5 ppt isopycnal. It is to be noted that at 0.5 ppt model response to the small fresh event in late 1994 is quite good although the seaward excursion is approximately 5 km too high.

Interpretation of movement of larger isopycnal values must be undertaken with due regard for this calibration procedure. Better agreement with higher isopycnals could be obtained by re-calibrating the model with the required isopycnal and re-running the simulation. However, these additional runs were considered to be outside the scope of the present study. WRL TECHNICAL REPORT 99/51 12.

2.8 Prediction of Salinity Structure

2.8.1 Empirical Calculations

One objective of this investigation is to predict changes to salinity structure of the Richmond River from freshwater extraction. It is appropriate that the irrigation extractions be set in the context of other losses of freshwater from the Richmond system.

If the mean water usage by irrigators is 2Ml/Ha/yr, the net outflow from the Richmond tidal pool is 620Ml/month. In the absence of freshwater inflow at the tidal limits, this would result in a landward movement of the salt/fresh water interface (assuming an 200 m wide by 4m deep average cross-section) of 800m per month.

By comparison, evaporation rates are typically 6 to 13 mm per day in coastal regions of Australia. For a river that is 40 km long and 80 m wide with a 6mm per day evaporation rate - this is a volume flux of fresh water greater than 576Ml/month. This is roughly equivalent to that currently removed from the system by irrigators (assuming 2Ml/Ha/yr). Please note that evaporation from the surface of the river has not been incorporated within the long term simulations.

From the observations obtained in 1994, salt water from the sea was observed to travel upstream at a rate of approx. 60km/six months: equivalent to 10 km per month.

On these calculations, irrigation is currently increasing the landward movement of saline waters by approximately 8%. Irrigation has been taking place on the Richmond for well over half a century. Increasing irrigation rates, will lead to increased landward movements of a saline water and the significance of these is discussed in Section 3.6.

It is also to be noted that the estuary reduces in width and depth with distance from the ocean. The calculations presented above have been made for a channel cross-section representative of the downstream reaches of the estuary. Consequently, landward transports of salt in terms of distance in a given time will increase away from the ocean.

2.8.2 Model Simulations and Management Scenarios

The model has been used to simulate the salinity characteristics of the Richmond River tidal pool for estimated freshwater inflows for the period 1940 to 1998.

Simulations have been undertaken for: WRL TECHNICAL REPORT 99/51 13.

• Existing Conditions: with the estimated freshwater inflows and freshwater extraction within the tidal pool corresponding to 2Ml/Ha/yr irrigation rates; • Option A: existing conditions with twice the level of freshwater extraction within the tidal pool; • Option B: existing conditions with 3.5 times the current level of freshwater extraction within the tidal pool; • Option C: the estimated freshwater inflows with no freshwater extraction within the tidal pool; • Option D: existing conditions with expected sea level rise for 2050; • Option E: existing conditions with 5 times the current level of freshwater extraction within the tidal pool; • Option F: existing conditions but with freshwater inflows reduced by 30%.

2.8.3 Discussion of Results

Existing Conditions Figures 2.22 to 2.24 show the modelled salt distributions for the assessment period. These results highlight the strong fluctuations in isopycnal position along the estuary on time scales of months to years.

Although 1994 was believed to be a dry year, within this context it is not exceptional as shown in Figure 2.16. The entire period from 1989 to 1995 shows fluctuations of saline intrusion of similar magnitude in each year.

Water Resources of the Richmond Valley states that "the lowest calendar year rainfall for Casino was 19.29 inches (490mm) in 1902 and the next lowest was 19.73 inches (501mm) in 1915. The longest sequence of below average falls occurred from 1940 to 1946 and the lowest 5 year period occurred from 1940 to 1944 inclusive."

This is reflected in the results in Figure 2.22 but saline intrusion to the estuary is similar for the periods 1940-1946, 1976-1982 and 1989-1995.

The model indicates that the most landward excursion of saline waters into the estuary was in 1970 to a distance of 90 km from the ocean.

It must be emphasised that geomorphologists regard the last 50 years as relatively wet in comparison with the first 50 years of this century. If long-term averages are to be WRL TECHNICAL REPORT 99/51 14.

considered, landward excursions of saline waters may be underestimated by this assessment.

Option A (Figures 2.25 to 2.27) If the level of freshwater extraction from the tidal pool is increased by a factor of two, the landward intrusion of salt water on the Richmond River and Bungawalbin Creek increases by approximately 2 to 5km during dry periods. The corresponding increase on the Wilsons River is between 4 and 7 km.

The duration of the dry periods is generally in the vicinity of six months. These values are in agreement with the estimates obtained from the empirical calculations.

Option B (Figures 2.28 to 2.30) If the level of freshwater extraction from the tidal pool is increased by a factor of 3.5 above existing levels, the landward intrusion of salt water increases by between 5 and 20 km during dry periods. For the 1970 dry period, saline waters can be observed to intrude up the Richmond River to the tidal limit at Casino and to the tidal limit of the Wilsons River and Bungawalbin Creek.

Option C (Figures 2.31 to 2.33) If the level of freshwater extraction from the tidal pool is decreased to zero, there is a corresponding decrease in the landward intrusion of salt water by between 2 and 5 km during dry periods.

Option D (Figures 2.34 to 2.36) For this option the tailwater level of the model was increased by 0.39 m, the upper estimate for sea level rise for 2050. The landward intrusion of sea water is increased. This is primarily because the freshwater flushing velocities in the channel are decreased. However, the differences from the existing condition are very small. Intrusion is perhaps 500m greater for this scenario.

It could be argued that the increased depths resulting from a rise in water levels would cause greater density intrusion and mixing and therefore greater intrusion into the estuary. Density effects cannot be included in this model.

Option E (Figures 2.37 to 2.39) If the level of freshwater extraction from the tidal pool is increased by a factor of 5 above existing levels, the landward intrusion of salt water in the Richmond River arm increases by between 5 and 30 km during dry periods. Under these conditions, results indicated WRL TECHNICAL REPORT 99/51 15.

that saline water would approach the tidal limit at Casino on at least seven occasions between 1940 and 1997. Similar trends are evident for Wilsons River and Bungawalbin Creek, with instances of saline water being within 2 km of the tidal limit occurring on at least 4 and 6 occasions respectively during the simulation period.

Option F (Figures 2.40 to 2.42) As expected, the impacts of decreasing freshwater flows by 30% include the presence of saline water in the upper reaches of the estuary on a more frequent basis and a reduced flushing ability, in comparison to existing condition results.

The results demonstrate that a reduction in freshwater inflows to the estuary produces salinity intrusion distances only slightly greater than existing conditions. For example, on the main Richmond River arm, saline water moved between 5-8km further inland. Similarly, on the Wilsons River arm, saline intrusion was 2-8 km greater due to decreased freshwater flows.

However, when viewed in light of the results for options A, B and E, it becomes apparent that the degree of salinity intrusion up the estuary is more sensitive to the magnitude of the irrigation rates during the drier, low flow periods.

2.9 Simulation of 1901-1902

To investigate impacts of extreme dry weather periods upon saline intrusion, an additional analysis was undertaken for the year of 1902. Water Resources o f the Richmond Valley states that "the lowest calendar year rainfall for Casino was 19.29 inches (490mm) in 1902 and the next lowest was 19.73 inches (501mm) in 1915". The aim of this analysis was to provide an indication of what impacts would result from current irrigation practices should a similar prolonged dry weather period occur in the future.

The 1902 dry spell Was simulated starting with the below average year of 1901. Using data from those daily rainfall gauges in the study catchment which had records during 1901-1902, the AWBM model was used to derive hydrological inputs for the RMA-2 and RMA-11 models. The RMA runs used the AWBM generated flows in addition to adopting the irrigation rate used for "existing conditions" (2ML/ha/yr). The hydrodynamic and subsequent water quality simulation were then completed for this period (1901-1902).

Figure 2-43 shows the modelled salt distributions in the Richmond River arm during 1902. In comparison with the salt distributions modelled for 1994, it can be seen that the 0.5 ppt WRL TECHNICAL REPORT 99/51 16.

isopycnal extends a further 4km upstream of the most inland point simulated for 1994. For 1902, this corresponds to a region upstream of the of the Richmond and Wilsons Rivers at Coraki.

Figure 2.22 shows that 1994 was not a particularly exceptionally dry year and suggests that the hydrological analysis undertaken for 1940 to 1997 is representative of the preceding 50 years. WRL TECHNICAL REPORT 99/51 17.

3. ANALYSIS OF HABITAT

3.1 Approach

Estuarine have a vast number of biotic (living) and abiotic (non-living) components and linkages. It is, therefore, never possible to make meaningful predictions concerning ecosystem health by considering just one facet of the ecosystem. For this reason a range of ecosystem facets were targeted. That is, the investigation took a multifaceted approach which attempts to account for the potentially high level of ecosystem complexity. In this context, three parallel investigations, targeting a range components of the estuarine ecosystem, were undertaken:

• Vegetation (Dalby-Ball and Sainty): Appendix A • Fish communities and their habitats (Bishop): Appendix B • Platypus (Grant): Appendix C

The primary objective of the work was to gather and analyse pertinent information necessary for assessing the potential impacts of predicted shifts in the estuary's salinity structure. This involved:

• detailing species occurrences, distributions and information on their habitats, particularly in the upper arms of the estuary

• the identification of high-value components, particularly those most vulnerable to shifts in salinity structure, and

• the identification of salinity-exposure characteristics which best typify transition conditions where freshwater biota are displaced by more salt-tolerant biota, or where freshwater habitat is degraded (emphasis being placed on understanding the limitations of the thresholds obtained and the complications involved in their use).

3.2 Estuarine Vegetation

Aquatic and riparian vegetation were included in the assessment because of their potential susceptibility to changes in salinity through:

Direct-impact mechanism: physiological stress, Indirect-impact mechanism: from salt-tolerant flora WRL TECHNICAL REPORT 99/51 18.

Plant communities are considered to be high-value components of estuarine ecosystems because they provide an important ecological role in:

(i) providing shelter and/or food to , fish and other vertebrates, (ii) stabilising water quality (nutrients), and (iii) stabilising banks of estuaries (riparian and emergent plants).

The full report on estuarine vegetation is given in Appendix A. A summary of the key findings from the study is given below:

• Details of aquatic macrophyte and riparian species presence and distribution in the Richmond River estuary are not available with the exception of a study in Emigrant Creek, a small lower arm of the estuary. Observations in the lower main channel of the estuary indicate that dominate the edge vegetation to ~35km upstream where Phragmites australis becomes dominant.

• A literature review of salinity tolerances of aquatic and riparian vegetation, revealed high variability in reported tolerance limits for some species. Salinity limits determined from experimental studies and field observations vary greatly between and within species. Tolerance limits of plants are suggested to be greater in systems where salinity fluctuates due to freshwater inputs, tides, etc.

• Information on the responses of aquatic plants and riparian vegetation to salinity is too variable to make generalised predictions for the Richmond River estuary. A range of responses is expected for different species. To maintain , salt sensitive species need to be identified and their tolerances tested under the specific conditions that occur in the Richmond River estuary.

• Field observations from freshwater in South Australia suggest a general limit of 4 ppt salinity for wide-ranging Australian freshwater macrophytes. However, some Australian macrophytes show declining condition when salinities exceed 1 ppt. It is recommended that salinity levels are not allowed to exceed lppt in areas that are currently freshwater. This is an interim measure. It should be reassessed once further information is gained regarding actual species presence, distribution and field responses to increased salinity.

• Published tolerance limits are available for a few riparian species, but the applicability of the results to the Richmond River estuary is uncertain. It could be expected that WRL TECHNICAL REPORT 99/51 19.

some riparian plants could decline in condition if their roots are exposed to soil with salinity >2ppt. Accordingly, it is recommended that soil salinity should be kept below 2ppt in areas that are currently unaffected by salt. As before, this is an interim measure. It should be reassessed once further information is gained regarding actual species presence, distribution and field responses to increased salinity.

• Phragmites australis is an edge species currently found in the lower reaches of the estuary where there is saline influence. Die-off may occur if salinity levels rise above this plant's tolerance limits. Obtaining definition on the tolerance limits is difficult given that reported tolerance limits range from 1.5 to 13 ppt salinity.

3.3 Fish Communities and Their Habitats

Fish communities were included in the assessment because of their potential susceptibility to changes in salinity through a range of possible mechanisms:

• Direct-impact mechanisms: physiological stress

• Indirect-impact mechanisms: competition and from salt-tolerant , salinity-mediated impacts on their food , loss of productive habitat arising from salinity-mediated impacts on plants.

Fish communities are considered to be high-value components of estuarine ecosystems for a range of reasons:

• fish have- an important ecological role as primary consumers () and secondary and tertiary consumers (), • fish have traditionally been used in assessments • fish have a high public-appreciation, recreational and commercial value • some taxa have a high conservation value (notably Eastern freshwater cod, Maccullochella ikea, and Oxleyan pygmy perch, Nannaperca oxleyana, which occur in the Richmond system, and are afforded legal protecting under the Amendment Act 1997)

The full report on fish communities and their habitats is given in Appendix B. A summary of the key findings from the study is given below: WRL TECHNICAL REPORT 99/51 20.

• There were clear indications that the Bungawalbin Creek arm of the estuary potentially contains extensive high-value physical habitat. Unfortunately, there are also indications that the value of this habitat has recently been reduced by deteriorating water quality. To obtain more definition and confidence in the identification of high- value habitats, habitat surveys along the upper arms of the estuary would need to be undertaken.

• The upper Richmond River arm of the estuary potentially has the greatest complement of high-value fauna given the additional records of the high-conservation-value Oxleyan pygmy perch and Eastern freshwater cod. These records need to be confirmed and this would only be possible through intensive, well-focussed sampling.

• There were many inconsistencies amongst and between indicative salinity tolerances derived from the literature/discussions and those derived from the analysis of available data along salinity gradients. These inconsistencies reflect the complexities involved in understanding the way in which salinity influences fish communities in the upper arms of estuaries. Given an understanding that the majority of Australian have a relatively short evolutionary history in freshwaters, there is likely to be a reduced importance of direct physiological impacts arising from increases in salinity. Instead more attention should generally be focussed on indirect impacts such as habitat degradation or competition and predation arising from species with markedly-high salinity tolerances.

• Six indicative salinity limits arose from the study. Given that they are based on 'best- available' information, they can be used as 'working' thresholds to be input into risk analyses which assess the implications of extracting particular volumes of freshwater. The risk-analysis process helps to identify characteristics of extractions (e.g. volume, rate, and timing) which should minimise resultant impacts. The process as described has the advantage that it attempts to utilise what information is available, or readily attainable, on individual estuaries and their most valuable and/or vulnerable components. Of fundamental importance, the process utilises estuary-specific information on known links between hydrology and basic structural properties of the ecosystem.

• Given the complexities of estuarine ecosystems and inaccuracies which may occur in the methodologies, particularly in relation to the range of working thresholds utilised, it is imperative that any implemented extraction regime be viewed as an interim condition, to be revised once substantial knowledge is gained through ensuing WRL TECHNICAL REPORT 99/51 21.

scientific research and monitoring. This is a fundamentally important feature of any adaptive management system.

• Monitoring in estuarine ecosystems is potentially more difficult than in river/stream ecosystems due to the complexities introduced by the daily tidal cycle. To enable comparisons between places and times, samples should be taken at the same phase of the tidal cycle, and the same time of day. Attempts to utilise comparable combinations of tidal phase, and time of day, will result in protracted sampling sessions.

3.4 The Platypus

The platypus, Ornithorhynchus anatinus, was included in the assessment because of its potential susceptibility to changes in salinity through a range of possible mechanisms:

• Direct-impact mechanisms: physiological stress (high-salt content foods and incidentally ingested water), reduced feeding efficiency due to lowered electrosensitivity, increased coat-maintenance problems,

• Indirect-impact mechanisms: competition and predation from salt-tolerant fauna, salinity-mediated impacts on their food organisms, loss of productive habitat arising from salinity-mediated impacts on plants.

The platypus also has high public-appreciation and conservation value.

The full report on the species is given in Appendix C. A summary of the key findings from the study is given below:

• Data from literature and database searches, including information from the Richmond River estuary, indicated that platypuses are largely restricted in their distribution to the upper sections of estuaries, within a few kilometres of tidal influence. These sections of the Richmond River estuary are essentially freshwater (< 0.5 ppt).

• The reasons for such a distribution is not known, but is probably related to a number and/or combination of biotic and abiotic factors influenced by tidal cycles in the estuarine environment, rather than to the species' tolerance to a specific level of a single factor, such as salinity.

• Encroachment of greater salinity further up the arms of the Richmond estuary, due to the extraction of water from tidal pools, may not directly affect the platypus in terms of WRL TECHNICAL REPORT 99/51 22.

its direct tolerance of higher saline conditions, but may displace whatever factors are determining the distribution of the species further upstream, thereby reducing its distribution within the system.

• It was recommended that:

any proposed extraction regimes be modelled as accurately as possible, an adaptive management strategy be developed, which includes strict monitoring programs and a facility to change the extraction regimes in response to either environmentally positive or negative findings of monitoring studies, and at least initially no extraction, or very limited extraction, be made from the smaller tidal pools, including Bungawalbin and Emigrant Creeks.

3.5 Indicative Salinity Limits and Their Use

A consolidated list of recommended indicative salinity limits arising from the three study facets is given below:

Estuarine water salinity:

0.12 ppt: upper limit for Oxleyan pygmy perch 0.5 ppt: upper limit for the platypus (indirect impacts) lppt: upper limit for Eastern freshwater cod (indirect impacts), eeltailed catfish (Tandanus sp.A), the maintenance of freshwater ecosystems, and aquatic macrophytes currently in freshwater areas 5 ppt: upper limit for adult Australian bass outside the spawning season as well as for ribbon grass which is important the shelter of larval and juvenile bass. 13 ppt: upper limit for adult Australian bass during the spawning season. 20 ppt: indicative limit for the Sydney rock oyster above which there is an increased chance of fouling of attachment substrates by marine animals. 25 ppt: upper limit for common reeds which provide shelter for recruiting larval and juvenile bass.

Soil salinity:

2 ppt: upper limit for riparian vegetation currently adjacent to freshwater areas

The discussions in Appendices A, B and C describe the complexities involved in understanding the way in which salinity influences biotic communities in the upper arms of WRL TECHNICAL REPORT 99/51 23.

estuaries. It also illustrates the potential weaknesses of any derived salinity limits. Nevertheless, there is a useful role for such limits if they are based on ’best-available' information. Being indicative of points on the salinity spectrum where biological changes may commence, they can be used as 'working' thresholds to be input into risk analyses which assess the implications of extracting particular volumes of freshwater.

The risk-analysis process helps to identify characteristics of extractions (e.g. volume, rate, and timing) which should minimise resultant impacts. Important here is the provision of a means to understand and demonstrate the nature of tradeoffs between increased water extraction, and increased impacts on the estuarine ecosystem. The process, as described in more detail in Appendix B, has the advantage that it attempts to utilise what information is available, or readily attainable, on individual estuaries and their most valuable and/or vulnerable components. Of fundamental importance, the process utilises estuary-specific information on known links between hydrology and basic structural properties of the ecosystem.

Given the complexities of estuarine ecosystems and inaccuracies which may occur in the methodologies, particularly in relation to the range of working thresholds utilised, it is imperative that any implemented extraction regime be viewed as an interim condition, to be revised once substantial knowledge is gained through ensuing scientific research and monitoring. This is a fundamentally important feature of any adaptive management system.

3.6 Risk Analysis of Water Extraction Scenarios

3.6.1 Dimensions of the Analysis

The analysis was designed to allow risk to be considered in relation to differences between:

• extraction options • major arms of the estuary • ecosystem facets • freshwater-content conditions of the estuary (i.e. the size of its tidal pool)

The extraction options are:

Option C: no extraction from the tidal pool, i.e. the ’natural' condition Existing: extraction arising from an application rate of 2 Ml/Ha/yr, i.e. the existing or current condition WRL TECHNICAL REPORT 99/51 24.

Option D: extraction arising from an application rate of 2Ml/Ha/yr with the addition of the effect of the sea level rising to the level expected by the year 2050 Option F: extraction arising from an application rate of 2Ml/Ha/yr with the addition of the effect of freshwater inflows reduced by 30% Option A: extraction arising from an application rate of 4Ml/Ha/yr Option B: extraction arising from an application rate of 7Ml/Ha/yr Option E: extraction arising from an application rate of lOMl/Ha/yr

The major arms of the estuary considered are:

• Richmond River (main) arm • Wilsons River arm • Bungawalbin Creek arm

The ecosystem facets are those identified in Section 3.5 above in relation to the six indicative estuarine-water salinity limits, or 'working' salinity thresholds. Ecosystem facet 7, as associated with the only soil-salinity threshold (relevant to riparian vegetation), is not included as the availability of habitat with suitable salinity cannot be reliably modelled in relation to the extraction options, a result of the complexity of water-soil interactions. Such modelling (see Chapter 2 for the background to the process), which is possible for the six facets associated with estuarine-water salinity limits, forms the basis of the risk-assessment process as described below in Sections 3.6.2 to 3.6.5.

The freshwater-content conditions o f the estuary considered are:

• 50th percentile condition: 1 ppt salinity reaching 24km upstream of the estuary mouth under the existing condition (UEMUEC), i.e. near Broadwater. • 80th percentile condition: 1 ppt salinity reaching 43 km UEMUEC, i.e. near Woodbum. • 90th percentile condition: 1 ppt salinity reaching 54km UEMUEC, i.e. near the Bungawalbin Creek confluence. • 95th percentile condition: 1 ppt salinity reaching 60km UEMUEC, i.e. near Coraki. • 99th percentile condition: 1 ppt salinity reaching 69 km UEMUEC.

These percentile conditions reflect an 'integration' of the recent history of freshwater inflows to the estuary. Given that the conditions with greater freshwater content (i.e. towards the 50th %ile condition) may arise from either a high-intensity/short-duration inflow event, or a low-intensity/long-duration event, it is inappropriate to associate the conditions with particular climatic conditions. Nevertheless, conditions with a much lower WRL TECHNICAL REPORT 99/51 25.

freshwater content, particularly the 99th %ile condition, would clearly be associated with drought conditions.

3.6.2 Habitat-Availability Data Arising from Salinity Thresholds

The modelling process described in Chapter 2 facilitated habitat-availability modelling as relevant to the six ecosystem facets, and as defined by their salinity thresholds.

The measure of habitat availability varied to some extent between ecosystem facets:

• Ecosystem facet 1 (Oxleyan pygmy perch):

length of estuary with salinity less than the threshold (i.e. length upstream of the point in the estuary where the threshold salinity concentration is predicted to occur)

length was considered to be an appropriate measure as the perch is expected to be an edge-dwelling species

• Ecosystem facet 2 (platypus):

length of estuary with salinity less than the threshold

length was considered to be an appropriate measure as platypus are strongly dependent on channel banks for sheltering and as breeding sites

• Ecosystem facet 3 (general + cod + catfish + macrophytes):

area of estuary with salinity less than the threshold (i.e. area upstream of the point in the estuary where the threshold salinity concentration is predicted to occur)

area was considered to be an appropriate measure as all four components would not be expected to be restricted to the edge zone of the channel

• Ecosystem facet 4 (non-spawning bass + ribbon grass):

area of estuary with salinity less than the threshold

area was considered to be an appropriate measure as both components would not be expected to be restricted to the edge zone

• Ecosystem facet 5 (spawning bass):

area of estuary with salinity less than the upper threshold, but greater than the lower threshold (i.e. the area between the points in the estuary where the upper and the lower threshold-salinity concentrations are predicted to occur) WRL TECHNICAL REPORT 99/51 26.

area was considered to be an appropriate measure as bass spawning would not be expected to be restricted to the edge zone of the channel

• Ecosystem facet 6 (Sydney rock oyster & marine fouling):

length of estuary with salinity less than the threshold within the interval where oysters are cultivated, i.e. from 2 to 10 km upstream of the estuary mouth

length was considered to be an appropriate measure as oyster cultivation is generally restricted to the edge zone

3.6.3 Assumptions Regarding Risk

Assumption 1

It was assumed that biota least exposed to variation in salinity are those most at risk from changes to the salinity structure of the estuary. Supporting this assumption, Montague and Ley (1993; reference given in Appendix A) indicated that a measure of variation in salinity levels was a better predictor of macrophyte distribution and than mean salinity.

Longitudinal changes in salinity percentiles along the main arm of the Richmond River estuary are given in Figure 3.1 for the existing condition. This figure shows that salinity variation is at a maximum about 5 km upstream of the estuary's mouth then steadily declines to a point ~25 km upstream. The reduction in variability then slows noticeably to reach very low levels in a section 55-65 km upstream. Following the assumption, it is expected that biota most at risk from changes to the salinity structure of the estuary are present within and upstream from this section. In a general sense this provides a spatial focus for the assessment.

Assumption 2

With the extraction of freshwaters from the upper arms of the estuary, saline waters will penetrate further upstream thus contracting habitat for the more freshwater-orientated biota. It is assumed that risk increases as the extent of the habitat contraction increases.

3.6.4 Measuring Habitat Contraction and Risk Categories

The extent of habitat contraction associated with each extraction option was calculated as the percentage change from that expected under the 'natural' condition. It was assumed that Option C was the best approximation for the 'natural' condition, i.e. no extraction of WRJL TECHNICAL REPORT 99/51 27.

freshwaters from the estuary. The percentage habitat change was calculated for each major arm of the estuary, each ecosystem facet, and for each percentile condition of the estuary.

Following consultation with other aquatic scientists, the following risk (R) categories were set prior to data analysis and interpretation:

negligible: R <= 1% loss low: 1% < R <= 10% loss moderate: 10% < R <= 20% loss high: 20% < R <= 40% loss very high: 40% < R

On a broad scale the category divisions encapsulate the concept that risk increases as the extent of habitat contraction increases. On a fine scale the actual placement of the category boundaries may be subject to some minor debate. However, the approach has the advantage that it is open to scrutiny, being able to be revised if needs be with alternative category boundaries.

3.6.5 Percentile Condition Synthesis and Thresholds

The results of the assessment are synthesised in relation to the percentile conditions of the estuary as this is likely to best tie in with future water management procedures. That is, restrictions on water extraction are likely to increase as the availability of freshwater within the estuary decreases.

The risk assessment focuses on the high and very high risk categories as these are considered to -be the key environmental consequences to be avoided. A risk index was developed in order to examine the nature of the relationship between risk and the water extraction rate. Of particular interest was the identification of the extraction rate after which high and very-high risk events commence, i.e. the identification of the extraction- rate threshold where the risk index commences to rise above the zero level. This threshold is henceforth referred to as the 'post-zero risk threshold'.

A simple gradient of extraction rates was obtained by examining changes in risk across the following extraction options, or extraction-rate surrogates:

Option C: 0 Ml/Ha/yr Existing: 2 Ml/Ha/yr Option A: 4 Ml/Ha/yr Option B: 7 Ml/Ha/yr Option E: 10 Ml/Ha/yr WRL TECHNICAL REPORT 99/51 28.

The risk index was devised to provide a greater weighting for very-high risk categories as opposed to the high risk categories. Additionally, when summing risk across ecosystem facets, greater weighting would be applied to facets which had more components.

The weightings for the risk categories were:

High risk: one Very-high risk: two

The weightings of the ecosystem facets was as follows:

Facets 1, 2, 5, 6: one (each contain 1 component) Facet 3: four (contains 4 components) Facet 4: two (contains 2 components)

Accordingly, the risk index (RI) was the sum of weighted scores across all ecosystem facets. It was determined for each combination of Option (as specified above), major estuary arm, and estuary-percentile condition. It was calculated as follows:

RI = (Hhi + 2Vei) + (He2 + 2Vh 2) + 4(He3 + 2Vh 3) + 2(He3 + 2VE3) + (HE5 + 2VE5) + (He6 + 2Ve6) where: H = number of high risk categories V = number of very high risk categories El to E6 = Ecosystem facets 1 to 6

3.6.6 Changes in Habitat Availability and Risk

Changes in the length/area of habitat remaining for each ecosystem facet in relation to the percentile conditions (% time exceedance), and the extraction options, are shown in the following figures:

Facet 1: Figure 3.2a,b,c Facet 2: Figure 3.3a,b,c Facet 3: Figure 3.4a,b,c Facet 4: Figure 3.5 a,b Facet 5: Figure 3.6 Facet 6: Figure 3.7 WRL TECHNICAL REPORT 99/51 29.

The 'a,b,c' divisions relate to separate figures for the Richmond, Wilsons and Bungawalbin arms of the estuary respectively. Figures 3.6 and 3.7 are for the Richmond arm of the estuary only.

A summary of the percentage change in the length/area of estuarine habitat with salinities suitable for the six ecosystem facets is given in Figure 3.8. The change is shown in relation to the extraction options, major arms of the estuary, and the percentile conditions of the estuary.

Ecosystem facet 1 (Oxleyan pygmy perch)

50th %ile condition.All options had risks classed low in the Richmond arm. All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

80th %ile condition. In the Richmond arm risks associated with the options ranged from low to high: only Option E had a risk classed as high. In the Wilsons arm risks for options ranged from negligible to very high: only Option B had a risk classed as high, and only Option E had a risk classed as very high. In the Bungawalbin arm risks for options ranged from moderate to very high: Options F and A had risks classed as high, and Options B and E had risks classed as very high.

90th %ile condition. In the Richmond arm risks associated with the options were classed in the same way as they were in the 80th %ile condition. In the Wilsons arm risks for options ranged from moderate to very high: Options F and A had risks classed as high, and Options B and. E had risks classed as very high. In the Bungawalbin arm risks for options ranged from high to very high: Options 'Existing', D, F and A had risks classed as high, and Options B and E had risks classed as very high.

95th %oile condition. In the Richmond arm risks associated with the options ranged from low to high: only Option B had a risk classed as high, and only Option E had a risk classed as very high. In the Wilsons arm risks for options ranged from moderate to very high: Options D, F and A had risks classed as high, and Options B and E had risks classed as very high. In the Bungawalbin arm risks for options were classed in the same way as they were in the 90th %ile condition.

99th %ile condition. In the Richmond arm risks associated with the options ranged from low to very high: Options F and A had risks classed as high, and Options B and E had risks classed as very high. In the Wilsons arm risks for options ranged from high to very high: WRL TECHNICAL REPORT 99/51 30.

Options 'Existing' and D had risks classed as high, and Options F, A, B and E had risks classed as very high. In the Bungawalbin arm risks for options ranged from high to very high: only Option 'Existing' had a risk classed as high, and all remaining options had risks classed as very high.

Ecosystem facet 2 (platypus)

50th %ile condition. All options had risks classed as either negligible or low in the Richmond arm. All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

80th %ile condition. In the Richmond arm risks associated with the options ranged from low to moderate. In the Wilsons arm risks for all options were classed as negligible (no change expected). In the Bungawalbin arm risks for options ranged from negligible to high: only Option E had a risk classed as high; all other options had risks classed as negligible (no change expected).

90th %ile condition. In the Richmond arm risks associated with the options ranged from low to high: only Option E had a risk classed as high. In the Wilsons arm risks for options ranged from negligible to very high: Options F and B had risks classed as high, and only Option E had a risk classed as very high; all other options had risks classed as negligible (no change expected). In the Bungawalbin arm risks for options ranged from low to very high: Options F and A had risks classed as high, and Options B and E had risks classed as very high.

95th %ile condition.In the Richmond arm risks associated with the options ranged from low to high: only Options B and E had risks classed as high. In both the Wilsons and Bungawalbin arms risks for options ranged from moderate to very high: Options F and A had risks classed as high, and Options B and E had risks classed as very high.

99th %ile condition. In the Richmond arm risks associated with the options ranged from low to high: only Option B had a risk classed as high, and only Option E had a risk classed as very high. In the Wilsons arm risks for options were the same as they were for the 95th %ile condition. In the Bungawalbin arm risks for options ranged from high to very high: Options 'Existing', D and F and A had risks classed as high, and Options B and E had risks classed as very high. WRL TECHNICAL REPORT 99/51 31.

Ecosystem facet 3 (general + cod + catfish + macrophytes)

50th %ile condition. All options had risks classed as low in the Richmond arm. All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

80th %oile condition. In the Richmond arm risks associated with the options ranged from low to high: Options F, B and E had risks classed as high. All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

90th %ile condition. In the Richmond arm risks associated with the options ranged from moderate to very high: Options F, A and B had risks classed as high, and only Option E had a risk classed as very high. In the Wilsons arm risks for options ranged from negligible to high: only Option E had a risk classed as high; all other options had risks classed as negligible (no change expected). In the Bungawalbin arm risks for options ranged from negligible to very high: only Option B had a risk classed as high, and only Option E had a risk classed as very high.

95th %ile condition. In the Richmond arm risks associated with the options ranged from moderate to very high: Options F and A had risks classed as high, and Options B and E had risks classed as very high. In the Wilsons arm risks for options ranged from negligible to very high: only Option B had a risk classed as high, and only Option E had a risk classed as very high. In the Bungawalbin arm risks for options ranged from moderate to very high: Options F and A had risks classed as high, and Options B and E had risks classed as very high.

99th %ile condition. In the Richmond arm risks associated with the options ranged from moderate to very high: Options F and A and B had risks classed as high, and only Option E had a risk classed as very high. In the Wilsons arm risks for options ranged from moderate to very high: only Option D had a risk classed as high, and Options F, A, B and E had risks classed as very high. In the Bungawalbin arm the risk classification for options was the same as it was for the 95th %ile condition.

Ecosystem facet 4 (non-spawning bass + ribbon grass)

50th %ile condition. All options had risks classed as low in the Richmond arm. All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected). WRL TECHNICAL REPORT 99/51 32.

80th %ile condition. In the Richmond arm risks associated with the options ranged from low to moderate. All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

90th %ile condition. In the Richmond arm risks associated with the options ranged from low to high: only options B and E had risks classed as high. All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

95th %oile condition. In the Richmond arm risks associated with the options ranged from low to very high: only Option B had a risk classed as high, and only Option E had a risk classed as very high. All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

99th %oile condition. In the Richmond arm risks associated with the options ranged from low to very high: Options F and A had risks classed as high, and Options B and E had risks classed as very high. Virtually all options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected). The exception was Option E in the Bungawalbin arm which had a risk classed as high.

Ecosystem facet 5 (spawning bass)

50th %oile condition. All options had risks classed as negligible in the Richmond arm (gains in spawning area were apparent, particularly with Option F, and to a lesser extent Option E). ,A11 options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

80th %ile condition. All options had risks classed as negligible in the Richmond arm (gains in spawning area were apparent). All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

90th %ile condition. All options had risks classed as negligible in the Richmond arm (gains in spawning area were apparent, particularly with Option E, and to a lesser extent with Option B). All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

95th and 99th %ile conditions. All options had risks classed as negligible in the Richmond arm (gains in spawning area were apparent, particularly with Option E, and to a lesser extent with Option B then Options F and A). All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected). WRL TECHNICAL REPORT 99/51 33.

Ecosystem facet 6 (Sydney rock oyster & marine fouling)

50th %ile condition. In the Richmond arm risks associated with the options ranged from low to high: only Option F had a risk classed as high (Options B and E both had risks classed as moderate). All options had risks classed as negligible in the Wilsons and Bungawalbin arms (no change was expected).

80th %ile condition. All options had risks classed as negligible in all arms of the estuary (no change was expected).

90th %ile condition. All options had risks classed as negligible in all arms of the estuary (no change was expected).

95th %oile condition. All options had risks classed as negligible in all arms of the estuary (no change was expected).

99th %ile condition.All options had risks classed as negligible in all arms of the estuary (no change was expected).

Key Patterns in Risk

The key patterns in risk are best revealed by the shifts in colour patterns on Figure 3.8, particularly the shift from 'cool' (blue, green) to 'hot' (red, yellow) colours which represents a shift from low risk to higher risk.

Some of the key patterns revealed in the analysis were:

• an increase in risk moving from the low extraction options (e.g. the existing condition) to the high options (e.g. Option E) • an increase in risk moving from the low to the high percentile conditions of the estuary arms, i.e. there is a greater risk of environmental damage from extraction when there is less freshwater in the estuary as occurs during drought conditions • an increase in risk moving from the Richmond, to the Wilsons, then to the Bungawalbin arm (greater impacts due extraction would be expected in the latter arm as it is the smallest arm); this is only apparent for the first three facets as their salinity limits occur within or close to the Wilsons and Bungawalbin arms under existing conditions • an increase in risk moving from the ecosystem facets with higher salinity limits to those with lower limits; this is most obvious within the Wilsons and Bungawalbin arms moving from Facet 3 down to Facet 1 WRL TECHNICAL REPORT 99/51 34.

3.6.7 Synthesis

Relationships between the risk index and the extraction rate surrogate are shown in Figure 3.9 for four percentile conditions of the estuary and for each major arm of the estuary.

50th percentile condition

This percentile condition is not shown as the risk index remains at zero in each arm of the estuary across all extraction rates examined in this section. Accordingly, extraction-rate surrogates at least up to and including lOMl/Ha/yr appear to be acceptable under this percentile condition.

It should be noted that extraction Option F (not examined on a relationship basis within this section) showed a high risk classification for Facet 6 (oysters & marine fouling) under this percentile condition within the Richmond arm of the estuary.

80th percentile condition

In both the Richmond and Wilsons arms of the estuary the post-zero risk threshold occurs between the 4 and 7 Ml/Ha/yr extraction-rate surrogates. Accordingly, extraction-rate surrogates at least up to and including 4 Ml/Ha/yr appear to be acceptable under this percentile condition in these arms.

In the Bungawalbin arm the post-zero risk threshold occurs between the 2 and 4 Ml/Ha/yr extraction-rate surrogates. Accordingly, extraction-rate surrogates at least up to and including 2 Ml/Ha/yr, i.e. the existing condition, appear to be acceptable under this percentile condition in this arm.

It should be noted that under this percentile condition extraction Option F (not examined on a relationship basis within this section) showed a high risk classification for:

• Facet 1 (Oxleyan pygmy perch) within the Bungawalbin arm • Facet 3 (ecosystem etc) within the Richmond arm

90th percentile condition

In both the Richmond and Wilsons arms of the estuary the post-zero risk threshold occurs between the 2 and 4 Ml/Ha/yr extraction-rate surrogates. Accordingly, extraction-rate WRL TECHNICAL REPORT 99/51 35.

surrogates at least up to and including 2 Ml/Ha/yr, i.e. the existing condition, appear to be acceptable under this percentile condition in these arms.

In the Bungawalbin arm the post-zero risk threshold occurs between the 0 and 2 Ml/Ha/yr extraction-rate surrogates. Accordingly, until the post-zero risk threshold is more accurately determined, it is advisable that only minimal extractions are allowed in the arm under this percentile condition. It is notable that the Bungawalbin arm of the estuary potentially contains extensive areas of high-value physical habitat. The high value nature of the arm is discussed and illustrated in Section 3.2.1 and Figure 24 within Appendix B.

It should be noted that under this percentile condition extraction Option F showed a high risk classification for:

• Facet 1 (Oxleyan pygmy perch) within the Bungawalbin arm • Facet 2 (platypus) within the Bungawalbin arm

• Facet 1 (Oxleyan pygmy perch) within the Wilsons arm • Facet 2 (platypus) within the Wilsons arm

• Facet 3 (ecosystem etc) within the Richmond arm

Further, it should be noted that under this percentile condition extraction Option D (not examined on a relationship basis within this section) showed a high risk classification for:

• Facet 1 (Oxleyan pygmy perch) within the Bungawalbin arm

95th percentile condition

In both the Richmond and Wilsons arms of the estuary the post-zero risk threshold occurs between the 2 and 4 Ml/Ha/yr extraction-rate surrogates. Accordingly, extraction-rate surrogates at least up to and including 2 Ml/Ha/yr, i.e. the existing condition, appear to be acceptable under this percentile condition in these arms.

In the Bungawalbin arm the post-zero risk threshold occurs between the 0 and 2 Ml/Ha/yr extraction-rate surrogates. Accordingly, until the post-zero risk threshold is more accurately determined, it is advisable that only minimal extractions are allowed in the arm under this percentile condition. As mentioned before, it is notable that the Bungawalbin arm of the estuary potentially contains extensive areas of high-value physical habitat.

It should be noted that under this percentile condition extraction Option F showed a high risk classification for: WRL TECHNICAL REPORT 99/51 36.

• Facet 1 (Oxleyan pygmy perch) within the Bungawalbin arm • Facet 2 (platypus) within the Bungawalbin arm • Facet 3 (ecosystem etc) within the Bungawalbin arm

• Facet 1 (Oxleyan pygmy perch) within the Wilsons arm • Facet 2 (platypus) within the Wilsons arm

• Facet 3 (ecosystem etc) within the Richmond arm

Further, it should be noted that under this percentile condition extraction Option D showed a high risk classification for:

• Facet 1 (Oxleyan pygmy perch) within the Bungawalbin arm • Facet 1 (Oxleyan pygmy perch) within the Wilsons arm

99th percentile condition

In the Richmond arm of the estuary the post-zero risk threshold occurs between the 2 and 4 Ml/Ha/yr extraction-rate surrogates. Accordingly, extraction-rate surrogates at least up to and including 2 Ml/Ha/yr, i.e. the existing condition, appear to be acceptable under this percentile condition in this arm.

In the Wilsons arm of the estuary the post-zero risk threshold occurs between the 0 and 2 Ml/Ha/yr extraction-rate surrogates. Accordingly, until the post-zero risk threshold is more accurately determined, it is advisable that only minimal extractions are allowed in the arm under this percentile condition.

In the Bungawalbin arm the post-zero risk threshold occurs between the 0 and 2 Ml/Ha/yr extraction-rate surrogates. Accordingly, until the post-zero risk threshold is more accurately determined, it is advisable that only minimal extractions are allowed in the arm under this percentile condition. As mentioned before, it is notable that the Bungawalbin arm of the estuary potentially contains extensive areas of high-value physical habitat.

It should be noted that under this percentile condition extraction Option F showed either a high risk or very high risk classification for:

• Facet 1 (Oxleyan pygmy perch) within the Bungawalbin arm • Facet 2 (platypus) within the Bungawalbin arm • Facet 3 (ecosystem etc) within the Bungawalbin arm WRL TECHNICAL REPORT 99/51 37.

• Facet 1 (Oxleyan pygmy perch) within the Wilsons arm • Facet 2 (platypus) within the Wilsons arm • Facet 3 (ecosystem etc) within the Wilsons arm

• Facet 1 (Oxleyan pygmy perch) within the Richmond arm • Facet 3 (ecosystem etc) within the Richmond arm • Facet 4 (bass non-spawning) within the Richmond arm

Further, it should be noted that under this percentile condition extraction Option D showed either a high risk or very high risk classification for:

• Facet 1 (Oxleyan pygmy perch) within the Bungawalbin arm • Facet 2 (platypus) within the Bungawalbin arm

• Facet 1 (Oxleyan pygmy perch) within the Wilsons arm • Facet 3 (ecosystem etc) within the Wilsons arm WRL TECHNICAL REPORT 99/51 38.

4. CONCLUSIONS AND RECOMMENDATIONS

A model of the fresh-saltwater interface in the Richmond River estuary system has been developed which incorporates three sub-models:

• a hydrological model that estimates runoff response to rainfall during both wet and dry periods; • an estuary flow model; and, • an estuary salt model.

These models have been calibrated against available data and used to estimate saline excursion into the estuary from 1940 to 1997. The numerical modelling of the movement of the fresh/saltwater interface has highlighted the large excursions in its motion in response to rainfall and sustained periods of dry weather. The salinity exposure values along the estuary have been determined.

The assumed existing irrigation rate in the Richmond catchment has been assumed to be 2Ml/Ha/year. Eliminating irrigation or doubling the irrigation rate will change the maximum intrusion distances by between 2 to 5 km on the Richmond River and in Bungawalbin Creek. On the Wilsons River the corresponding distances are between 4 and 7 km.

If this irrigation rate is increase by a factor of 3.5, intrusion distances during dry periods increase by between 5 and 20 km. For many dry periods, saline waters penetrate to the headwaters of each estuarine arm. Accordingly, increasing the irrigation rate by a factor of 5 causes saline water to reach the tidal limits more frequently, with intrusion distances of up to 30km occurring during dry periods.

Saline intrusion is estimated to increase marginally due to sea level rise.

The magnitude and frequency of saline intrusion into the estuary arms were found to increase (in comparison to existing conditions) when freshwater inflows to the estuary were reduced by 30%. However, the magnitude of the intrusion was found to be generally less than that produced by increasing the irrigation rates beyond a factor of two - this feature was more predominant during the drier, low flow periods.

An assessment of the dryest year on record (1902) has shown that saltwater intrusion during 1902 was less than other periods during the last 50 years. The assessment also indicates that the simulation period (1940 -1997) is representative of the last century. WRL TECHNICAL REPORT 99/51 39.

It is to be noted that this analysis is based on assumed levels of irrigation within the tidal pool area and these conclusions may need to be reviewed once a more detailed assessment of water usage is complete. Freshwater inflows from the catchment above have been related to rainfall and it has been impossible to distinguish the reduction in inflow occurring due to irrigation above the tidal limit. Assessment of the changes in irrigation volumes upstream during dry periods should be considered.

Using the historical analysis of saltwater intrusion, three parallel investigations, targeting vegetation, fish and platypus were undertaken. The primary objective of the work was to gather and analyse pertinent information necessary for assessing the potential impacts of predicted shifts in the estuary's salinity structure. There are considerable uncertainties regarding the response of many freshwater biota to increased salinity levels.

Our investigations suggest that soil salinity is a critical issue for estuarine riparian vegetation and that there is little available information on the flow of saline estuarine waters into bed .

Based on the available data, a systematic classification of ecosystem risk was composed in relation to expected changes in the salinity of estuarine waters. Ecosystem facets were defined in terms of categories of biota dependent on particular salinity levels and physical characteristics of the estuary (for example, length or surface area). A risk index was created which focussed on high risk and very high risk changes to the defined ecosystem facets. Each facet was weighted according to the number of represented biota groups.

In summary, any changes in extraction rate on the Bungawalbin Creek arm are reflected in changes in ecosystem risk. This indicates that the Bungawalbin arm is already at risk and suggests that existing water usage should be reviewed. Very careful investigation is required before increases in freshwater extraction are contemplated. This is particularly important as the Bungawalbin arm was identified as containing extensive areas of high- value physical habitat.

The analysis suggests that existing irrigation rates have not significantly increased ecosystem risk within the Richmond River arm. However, if this rate is doubled, the risk does increase. This suggests that existing irrigation rates are acceptable but should only be increased with caution and after detailed investigation. WRL TECHNICAL REPORT 99/51 40.

Slight increases in ecosystem risk were identified between the no irrigation and existing conditions on the Wilsons River arm. Risk was noticeable only during extremely dry periods. Existing irrigation rates are probably acceptable but should only be increased if more detailed investigations show that this does not have an adverse ecological impact. Consideration could be given to eliminating irrigation during extremely dry periods.

Sea level rise showed only minor changes to ecosystem risk from existing conditions.

However, strong changes in ecosystem risk occur with reductions to the existing freshwater inflows at the tidal limits. It is important that recognition be given to the impact of freshwater extraction upstream of the tidal limit on the ecology of the estuary downstream. Also, the competing irrigation interests between the tidal pool and the catchment above the tidal limit should be recognised.

This investigation has examined the impacts of reduced freshwater flows on the salinity structure and its consequences for the Richmond River ecosystem. Reducing the freshwater flows will also have impacts on the general water quality of the system. This issue is important, especially where increased flows with high nutrient content are contemplated, but its consideration is beyond the scope of this investigation. WRL TECHNICAL REPORT 99/51 41.

5. REFERENCES

Boughton WC (1993), "A hydrograph based model for estimating the water yield of ungauged catchments", Inst Engs Aust, Nat Conf Publ 93/14, pp 317-324.

Boughton WC and Carrol DG (1993), "A simple combined water balance and flood hydrograph model", Inst Engs Aust, Nat Conf Publ 93/14, pp 299-304.

Water Conservation & Irrigation Commission (1966), Water Resources of the Richmond Valley, Survey of Thirty NSW River Valleys, Report No.2, May 1996.

Bishop KA (1995), "Hastings district water supply augmentation: freshwater ecology study for the intake pump station upgrading at Koree Island", Report prepared for Connell Wagner Pty Ltd on behalf of the NSW Public Works and Services.

Bishop KA (1998), "The Pacific Highway upgrade in the vicinity of Ballina, NSW: Eight- part tests on three threatened fish species", Report prepared for Connell Wagner Pty Ltd on behalf of the NSW Roads and Traffic Authority.

Bishop KA. (in prep.), "The Pacific Highway upgrade in the vicinity of Ballina, NSW: Species Impact Statements on three threatened fish species", Report prepared for Connell Wagner Pty Ltd on behalf of the NSW Roads and Traffic Authority.

Elder JW (1959), "The dispersion of marked fluid in turbulent shear flow", J. Fluid Mech. 5,554-560

King IP (1994), RMA-2 Version 6.5 Two-Dimensional Finite Element Model for Flow in Estuaries and , Resource Management Associates.

MHL (1978), "Richmond River: Tidal Monitoring 26 May 1977", Manly Hydraulics Laboratory Report 245.

MHL (1995), "Richmond River Estuary Data Collection - November 1994", Manly Hydraulics Laboratory Report 708.

MHL (1998), "Freshwater Extraction from the Richmond River Below the Tidal Limit", Manly Hydraulics Laboratory Report 937 prepared for the Department of Land and Water Conservation. WRL TECHNICAL REPORT 99/51 42.

Pollard DA (1993), " Report. Part B: Effects of Structural Flood Mitigation Works on Fish Communities", NSW Fisheries Research Internal Report, December 1993.

Pollard, DA and Growns 10 (1993), The Fish and Fisheries of the Hawkesbury- System, with Particular Reference to the Environmental Effects of the Water Board's Activities on this System", Interim Report to Sydney Water, April 1993.

West RJ (1993a), Estuarine Fisheries Resources o f Two South Eastern Australian Rivers, PhD Thesis, University of New South Wales.

West RJ (1993b), "Northern Rivers Report. Part A: Estuarine Fisheries Resources", NSW Fisheries Research Internal Report, December 1993.

West RJ and King RJ (1996), "Marine, Brackish, and Freshwater Fish Communities in the Vegetated and Bare Shallows of an Australian Coastal River", Estuaries, 19, 31-41.

PRECIP EVAP

\ \ SURFACE RUNOFF= EXCESS (1-BFI)*EXCESS r ~ ~ 7 ~ r SURFACE RUNOFF= s s (1-KS)*SS

BASEFLOW RECHARGE= BFI*EXCESS

BASEFLOW= (1-K)*BS

Unisearch Figure WRL SCHEMATIC REPRESENTATION OF THE AWBM MODEL 2.1

Report No. 99/51 F ig2-1.cdr

______

(uilu) Mjdaa ijejmey uee|/\| 1/01/94 1/02/94 1/03/94 1/04/94 1/05/94 1/06/94 1/07/94 1/08/94 1/09/94 1/10/94 1/11/94 1/12/94

(lulu) i|)dea jjounu

Unisearch AWBM MODEL VERIFICATION: Figure WRL RICHMOND RIVER AT CASINO 2.7

Report No. 99/51 Fig2-7.cdr (lulu) ij)dea nejmey ueeyy 1/01/94 1/02/94 1/03/94 1/04/94 1/05/94 1/06/94 1/07/94 1/08/94 1/09/94 1/10/94 1/11/94 1/12/94

(lulu) ijidea jjounu

Unisearch AWBM MODEL VERIFICATION: Figure WRL LEYCESTER CREEK AT ROCK VALLEY 2.8 Report No. 99/51 F ig 2 -S .c d r ( l u l u ) LM dea n ejm ey uee|/\| 1/01/94 1/02/94 1/03/94 1/04/94 1/05/94 1/06/94 1/07/94 1/08/94 1/09/94 1/10/94 1/11/94 1/12/94

o

(luui) Mjdea jjouny

Unisearch AWBM MODEL VERIFICATION: Figure WRL WILSONS RIVER AT ELTHAM 2.9 Report No. 99/51 Fig2-9.odr

80,000,000

k 70.000. 000 \ 1e 60,000,000 Q X 1 50,000,000 2cs \ 1------1 3 > 40,000,000 \ I < ■o | 30,000,000 — ______— £ U S | 20,000,000 £

10.000. 000

0 ______10,000 20,000 30,000 40,000 50,000 60,000 70, 300 80, 300 90, 300 100 000 110 000 Distance from River Mouth (m)

?nnnnnm

lannnrxn - \

\ ifinno000 - Q X ^ 14noo000 - o m § 1? 000 000 -J \ < 0) i & g 10000 000 - ______CO

■o c ... V. c

f

0 10,C00 20,000 30,000 40,000 50,000 60,000 70,000 80,000 90,000 100,000 110,000 Distance from River Mouth (m)

Unisearch Figure WRL CHANNEL VOLUME AND AREA: RICHMOND RIVER 2.12 Report No. 99/51 Fig2-12cdr Unisearch Figure WRL CHANNEL VOLUME AND AREA: WILSONS RIVER 2 -1 3 .Report No. 99/51

MESH DATA Number of Elements: 324 Number of Nodes: 649

120000 n | i i i—' T i i—i i |—m—n —| i i i—i—pi—r~r 118000

116000

114000

112000

110000

108000

106000

104000

102000

100000

98000

96000 f—

94000

92000

90000 —

88000 ~

86000

84000 \—

82000 I— 80000 -

78000 -

76000 —

74000 —

72000 —

70000 LJ j i I I I I I I I I

EASTING

NOTE To convert to AMG66 co-ordinates: add 6700000m to Northings : add 500000m to Eastings Unisearch Figure WRL FINITE ELEMENT MESH FOR RICHMOND RIVER ESTUARY 2.15

.Report No. 99/51 Fig2-15 cdr 1 •u■s □ Source: MHL Report 937, ReportNSWMHL 937, Dept. Source: Public Of andWorks Services

Unisearch WRL LOCATIONS OF CALIBRATION DATA FOR RMA-2 CALIBRATION: 3 NOVEMBER 1994 Report No. 99/51 T-(\|(1^lfl(OSCOO) 9 9 9 9 9 9 9 9 9 (QHVw) |3A0i J0JBM

Figure Unisearch COMPARISON OF MEASURED AND MODELLED TIDE LEVELS: WRL RICHMOND RIVER - 3 NOVEMBER 1994 2.17 .Report No. 99/51 Ftg2-17.cdr o o o o■'T

CD ■o o o o ino < CO a: CD C JH Cl o o o o ■O o P CO 3 CO Q •a 0 0 o Z "OO O I—, CD inO o 0 CO CM C n < 0 O o: E £ ’ c ■o 0 o c o> O co 0 L. CO .E 9> o « 3 o .c DC CO o O ^ 0 o w ■a oz "0 i | E 0 ■o S 1 o 1 S 0d 0 o o _o o a) c CO o S o a in -S' CO T3 'C 0 aCO oE O o L! o 0 o > o (E c o CO

0 o o CD o Q in 2 0 Ll

O)0DN(Dlfi^C1Ni-Oi-(NM^in(DN000) ooooooooo oo ooooooo (CJHVW) |3A0-| J0JBM

Figure Unisearch COMPARISON OF MEASURED AND MODELLED TIDE LEVELS: WRL WILSONS RIVER - 3 NOVEMBER 1994 2.18 Report No. 99/51 Fig2-18.cdr

Wilson River: Comparison of Model and Observed Data: High Tide Lag

RMA Model Chainage (m) (from Richmond River confluence)

Wilson River: Comparison of Model and Observed Data: Low Tide Lag

RMA Model Chainage (m (from Richmond River confluence)

Figure Unisearch COMPARISON OF MEASURED AND MODELLED TIDAL LAGS: WRL WILSONS RIVER - 3 NOVEMBER 1994 2.20 import No. 99/51 P'92-20 cdr NteiF&rfd (mr} VbaiFlcwC^lh(rrm) o o o o o o o (N CO 00 CM t— O 1-Jan-95 eot o 99/51 No. Report

Unisearch 0.2 ppt WRL EXISTING CASE: MODELLED SALT DISTRIBUTIONS SALT MODELLED CASE: EXISTING NTERCMN RVR 1940-1997 RIVER RICHMOND THE IN (l u) L|inO|AI u) 0 J 0 OUBJSjQ Figure 22 .2 2 Rfl2-22.cdr Ow {M a r-o oN

3A-0ec-®

— A-Jafv*

3A-0et>®

A-J3>®

3A-Dec-49

n — i------1— r i— i------1 i i i r A-Jan-40

S3" S 8 S S3 8 u o O LO (ill) q p v \l 0190UB1SIQ

Unisearch Figure EXISTING CONDITION: MODELLED SALT DISTRIBUTIONS 2.23 WRL IN THE WILSONS RIVER 1940-1997 Report No. 99/51 Fip2-23-cdr eot o 99/51 No. Report Unisearch 0.2 ppt ,5?a B WRL . . § S o 8 8 4 c 8 —)

J __ EXISTING CONDITION: MODELLED SALT DISTRIBUTIONS SALT MODELLED CONDITION: EXISTING I __ Si f^ 8 I _____

i i — [ - i i I __ L ------8 Q 8 N UGWLI CEK 1940-1997 CREEK BUNGAWALBIN IN i i i i i i i i i r i i i i i i i i i i i i i i 1 ------i- i i 1 ( lu ) ) o 8 8 iflnq/\| J9Aiy CM o 8 LO 0 } souejsiQ o 8 8 I'll o L O o“ o i ---- 8 o L O r J __ I __ I __ A-jav®— I _ _ Figure 1-Jan-4° 3A430O 2.24 A-Jav® 3A-Oeo^9 Fig2-24.cdr 3 ^000-® 49 -Oec-8©

3AX)ec^9

3

8 8 S 8 8 ^ 8 8

(uj ) Ljinq/\| j0Ajy 0] eouejsiQ

Unisearch Figure OPTION A: MODELLED SALT DISTRIBUTIONS 2.25 WRL IN THE RICHMOND RIVER 1940-1997 Report No. 99/51 Report Unisearch 0.2 pp* WRL * *28 11 o 99/51 No. f LO“ ® £ 8 t f 8 8 § 3 5f i i i— i n ------—r 1— n OPTION A: MODELLED SALT DISTRIBUTIONS SALT MODELLED A: OPTION ------NTEWLOS IE 1940-1997 RIVER WILSONS THE IN i i i i 1 (ill) Lj]n (ill) ------i i i i 1 01 i ' i ' /\| J /\| 8 Ajy Ajy 0 ) ) 0 OUBJSIC] T I I I I~T I I 3A-0eo*9 _ o Figure 2.26 g2- cdr d .c 6 -2 2 ig F U**® 3 A-Jaw€0 3 A-Oeo49 VDec-89

eot o 99/51 No. Report

Unisearch 0.2 ppt WRL OPTION B: MODELLED SALT DISTRIBUTIONS SALT MODELLED B: OPTION N H RCMN RVR 1940-1997 RIVER RICHMOND THE IN ( ui) L|]no|/\| J 9 /\iy 0 } aouejsjQ Figure 2.28 g2- c* ~ * .c 8 -2 2 tg F

| 1 1 1 1 ' £ ft § 1 i M & O C • l o cm t - J_____ — L

L

\------______1------j - 31-Oec-®

r 1 L

-

- A-Jafv^O “i i i—r ~ 1 T J T 1 I T r l — i—i i r~' 1 1 1 1 1 1 i 1 c :

~ urr czr ^ ff: $ ** 5 l $ ; § 1

flnq/V j0Ajy 0 } 9CXJB]S jQ M i

Unisearch Figure OPTION B: MODELLED SALT DISTRIBUTIONS 2.29 WRL IN THE WILSONS RIVER 1940-1997 .Report No. 99/51 Fg2-29.cdr 0 .2 pp* 3A-OGC-80

3A-0ec-®

A-3a^

3A-O0C-49

A-ster40

(tu) gro^j BAjy o } aouBjsjQ

Unisearch Figure OPTION C: MODELLED SALT DISTRIBUTIONS 2.31 WRL IN THE RICHMOND RIVER 1940-1997 Report No. 99/51 J_J__ I__ L

3A-Dec-88

— A-Jarv®

3A-0ec^

A-JarvCO

tt-Oec^9

T~"—i— i— i— i------1— i— i— i i i i i i i i r 4- A-javtf)

a i f a a a a UO O LO (m) ip y \l J0Ajy 0} aouB^sia

Unisearch Figure OPTION C: MODELLED SALT DISTRIBUTIONS 2.32 WRL IN THE WILSONS RIVER 1940-1997 Report No. 99/51 F g 2 - 3 2 c d r A-0ecr€B Jsr'40 3 vjs*v8) 2.33 3 VJao^ Figure 3A-0eo49 v l_ _ I _ LO III III II _ J O —r 1 LO --- i—i 8 i i i i r 1 (oi) mno^j JOAfcj OOUejSjQ 01 (oi) mno^j JOAfcj LI I I I l l j I i i i i i , i i i i I j l l I I I LI IN IN BUNGAWALBIN CREEK 1940-1997 ---- OPTION C: MODELLED SALT DISTRIBUTIONS § 8 8 i-l—i I I I I I I I I I i-l—i II I 8 WRL

ldd 2 0 Unisearch Report Report No. 99/51 8 8 8' J5 8' 8 8 8 8 8

(lu) L|inc^\| jaA y o i ooubjsq

Unisearch Figure OPTION D: MODELLED SALT DISTRIBUTIONS 2.34 WRL IN THE RICHMOND RIVER 1940-1997 Report No. 99/51 31-Oec-®

A-Jarv€D

3A-0ec-®

A-Jatv€D

3A-0ec49

A-Jao40

(lu) gjnoiAl -BAfcd °J sou^sjQ

Unisearch Figure OPTION D: MODELLED SALT DISTRIBUTIONS 2.35 WRL IN THE WILSONS RIVER 1940-1997 Report No. 99/51

Report

U 0.2 pf* nisearch WRL o 99/51 No. OPTION E: MODELLED SALT DISTRIBUTIONS SALT MODELLED E: OPTION NTE IHOD IE 1940-1997 RIVER RICHMOND THE IN (tu) qinoiAi j9Aiy \ o eouejsjo Figure 2.37 37. r d .c 7 -3 2 g R j=--‘ 1 i ' 1 1 X 1 14-1 i-i i ■ -* ' i i i i i i i i i i i

<5 o © 3 N ^ ... L .. | .. [

j:______■ UL- j —

3A*0ec-$*

I-

-

-

-

- 31-Dec-eS

-

I l I I I I T I I I I I | 'I i r~!“ 1 1 1 1 1 1 1 1 1 i i 1 1 1 1 i i i i 1 1 I I - A-Ja^40 s £ 8 £ £ £ cz c i >l £c c£ c o c c: c: c ) lO c: LTr c uz' 5 5F c^3 r sp s CNJ C\ (W)l\p c \N j0Ajy 0 } e o u e ]SjQ

Unisearch Figure OPTION E: MODELLED SALT DISTRIBUTIONS 2.38 WRL IN THE WILSONS RIVER 1940-1997 Report No. 99/51 F g 2 -3 8 .c d r

(iu) iflnqfl J0Aiyo\ aoiEjsja

Unisearch Figure OPTION F: MODELLED SALT DISTRIBUTIONS 2.40 WRL IN THE RICHMOND RIVER 1940-1997 import No. 99/51 Fig2-40.cdr No. S. S o o. Q o. O CM ™

3A-0et>se

VJ*v®

StSeO©

-h A-Jao®

3A-O0^9

A-Jao40

8" 3" Q 8 « » R (w) ijinq/\| J0Aiyo\ aouejSjQ

Unisearch Figure OPTION F: MODELLED SALT DISTRIBUTIONS 2.41 WRL IN THE WILSONS RIVER 1940-1997 Report No. 99/51 ©o t o 99/51 “©port No. p KL L .K ” Uniseareli Uniseareli 0.2 ppt 8 o 8 LO o 8 c? o 8 PINF MODELLEDF:OPTIONSALT DISTRIBUTIONS 8 o 8 INBUNGAWALBIN1940-1997 CREEK ( lu ) 8 o 8 ninq/M jqa CN o 8 LO ^ j

______o \ aoLejsiQ 8 o 8 o 8 LO o' o 8 8 o LO o F A-jav40 9 4 o e 0 - A 3 A-Jarv^D 3A-0eo€S 2.42 A-Jarv® 3 ig2-42.cdr F A-Oec-S9 igure 1-Jan-03

______1'Jan'02

Disfcanoe From River M xith (m)

Unisearch Figure MODELLED SALT DISTRIBUTIONS 2.43 WRL IN THE RICHMOND RIVER, 1902. Report No. 99/51 Fig2-43.cdr ©

_c 4—» O h-

4 __ > I Z5 O I E

© CO s

Z5 4 —1 O CO lO CD E o

E © j*: S JZ 4-J O 0 oo © O p § c £ i© CO O © CD E GO © © Q _£Z _C © CO O a? £ ©> H £- © oc © © 5 o o o D CD cr >> © -Q I I © I I o f f © T3 I I E — © ^ & © _C Q_ CL X CL Cl O © © lO © © © r r (D h- I- Q

Unisearch Figure CHANGES IN SALINITY PERCENTILES ALONG THE MAIN ARM 3.1 WRL OF THE RICHMOND RIVER ESTUARY: EXISTING CONDITION Report No. 99/51 99069-3-1.cdr o o

o 4Q_ - * O o CD *

0 O c GQ LL O 4-» 00 CO Q_ Q. T30 O O 0 [3 O + 0X 0 O E < LU

Q_ CL O O

o CO

CD C -4—I Q CO X Q. LU O

O ID

Unisearch LENGTH OF HABITAT REMAINING FOR OXLEYAN PYGMY PERCH Figure WRL (THRESH. = 0.12 PPT) UNDER EXISTING CONDITIONS 3.2A

.Report No. 99/51 AND 6 OPTIONS 99069-3-2A.cdr o •4—* CL O

*

CD LL •4—• CL CL O o

CJ +

< LU 4-^ 4—' Q. CL O O

< 1 < >

o> c Q CO ^ X o . LU O

>k

Unisearch LENGTH OF HABITAT REMAINING FOR OXLEYAN PYGMY PERCH Figure WRL (THRESH. = 0.12 PPT) UNDER EXISTING CONDITIONS 3.2B

Report No. 99/51 AND 6 OPTIONS 99069-3-2B.cdr o o

o Q_ o o a>

0 O C CD LL o +-* -f-' oo CO CL Q. “O O O 0 0 O □ 4 * X 0 0 o K. e < LU +-> +-J Q . Q . o o

O co CDc Cl CO x Q- LU O

O uo

Uni search LENGTH OF HABITAT REMAINING FOR OXLEYAN PYGMY PERCH Figure WRL (THRESH = 0.12 PPT) UNDER EXISTING CONDITIONS 3.2C Report No. 99/51 AND 6 OPTIONS o o

o 4—» CL O o m *

0 o c CQ LL o -f—» -f—' 00 CD CL CL "O O O 0 0 □ O + X 0 0 o N- E < LU CL CL O o

o CD o> c 4—* Q CO X Q- UJ O

O to

Unisearch LENGTH OF HABITAT REMAINING FOR THE PLATYPUS Figure WRL (THRESH. = 0.5 PPT) UNDER EXISTING CONDITIONS 3.3A

Report No. 99/51 AND 6 OPTIONS 99069-3-3A.cdr Opt.E -+ - Opt.F Opt.D Opt.D

Unisearch LENGTH OF HABITAT REMAINING FOR THE PLATYPUS Figure WRL (THRESH. = 0.5 PPT) UNDER EXISTING CONDITIONS 3.3B

.Report No. 99/51 AND 6 OPTIONS 99069-3-3B.cdr o o

q CL O o G) *

CD O m LL O C 4—* ■f—• CL CL "O o O 0 0 □ o 4 - X 0 0 2 . i 4 - * < LU - f - j CL CL o o

o CD

cCD -f —' Q CO 4—> X CL LU O

o lO

Unisearch LENGTH OF HABITAT REMAINING FOR THE PLATYPUS Figure WRL (THRESH. = 0.5 PPT) UNDER EXISTING CONDITIONS 3.3C Report No. 99/51 AND 6 OPTIONS o Q_ O

*

m LL -i—» » Q. CL O O

C] +

LLJ

LLI — - 00 e £ CD <£> CD + - > Q) C CO +-» Q > X II CD -4—* 8 R X Q. LLI O g w JJ ® o X <0 O $ u . jC CD CO CD CO

ll CD LU CD LL ^

Unisearch AREA OF FE+EFC+ETC+MACROPHYTES REMAINING Figure WRL (THRESH. = 1 PPT) UNDER EXISTING CONDITIONS 3.4 A

Report No. 99/51 AND 6 OPTIONS 99069-3-4A. cdr o Q. O

I *

m u_ «♦—* -f—» Q. Q. O O

C] +

< LU -f-i CL Cl o o

<\ <>

CD C +-> Q CO X CL LU o

X FE= FE= , EFOfreshwater Eastern cod, ETC= eeltailed catfish,

Unisearch AREA OF FE+EFC+ETC+MACROPHYTES REMAINING Figure WRL (THRESH. = 1 PPT) UNDER EXISTING CONDITIONS 3.4B AND 6 OPTIONS Report No. 99/51 99069-3-4B.cdr o 4~1 Cl O

CD LL 4—> +-» CL CL O O

[] 4-

< LU 4—* 4—* Q. CL 4~*CD j f O O (O W cd m— LU <> cd <1 it O o ■D UL CD LU - CO

CD -Js§ Scd C CO ■4—> O >> II S2 ^ 8 R LU O 8 W l- TJ ® O i

Unisearch AREA OF FE+EFC+ETC+MACROPHYTES REMAINING Figure WRL (THRESH. = 1 PPT) UNDER EXISTING CONDITIONS 3.4C

Report No. 99/51 AND 6 OPTIONS 99069-3-4C.cdr o

CL o

m l l ■4—* «♦—* CL CL o o

t j 4

< UJ CL Q. o o

Q O CO LL1 o> CL c +j Q CD CO z CO x Q- < LU O z oc CD o: z COI o z ^m O 99 Z CO II II CO CD Z CO

Unisearch AREA OF HABITAT REMAINING FOR ADULT BASS NS AND RG Figure WRL (THRESH. = 5 PPT) UNDER EXISTING CONDITIONS 3.5A

.Report No. 99/51 AND 6 OPTIONS 99069-3-5A.cdr o Q_ o

CD LL •4—» Q. Q_ o O

u j

< LU Q . CL o o

<] Q o DC LU o> c CL •M q 0 CO CO Z CO x Q. — < LU O CE 0 a : Z * CO i O z m O m z DC ii ii CO o z cr

Unisearch AREA OF HABITAT REMAINING FOR ADULT BASS NS AND RG Figure WRL (THRESH. = 5 PPT) UNDER EXISTING CONDITIONS 3.5B

.Report No. 99/51 AND 6 OPTIONS 99069-3-5B.cdr o o

o oCL o CD %

0 O GQ LL O c ■M 00 CO Q. Q. u O O 0 0 [] O 4- X 0 0 O N- E < LLj 4—» CL CL o O

<3 <>

o CO CO c 4—* Q CO .+. * X Q_ LLi O

>k o I 1 0

Uni search AREA OF HABITAT REMAINING FOR BASS SPAWNING Figure WRL (THRESH = BT. 8-13 PPT) UNDER EXISTING CONDITIONS 3.6

Report No. 99/51 AND 6 OPTIONS 99069-3-6.cdr 2 0 3 _J cc O

Unisearch LENGTH OF HABITAT REMAINING FOR SR OYSTERS Figure WRL WITH LMF (THRESH. = <20 PPT) UNDER EXISTING CONDITIONS 3.7

Report No. 99/51 AND 6 OPTIONS 99069-3-7.cdr See text for details of the facets. The % change data are separated in relation to: * the existing condition and five extraction options (see text); the frame of reference of change is the condition with no extraction, i.e. Option C * the three major arms of the estuary * five percentile conditions indicative of reductions in the volume of freshwater present in the estuary, i.e. 50th%ile to 99th%ile (see text)

Key to colour codes used for risk categories (% change ranges are also given within the blocks):

Negligible Low Moderate High Very high risk or gain r i s k r i s k r i s k r i s k

+------+ + ------+ + ------+ +------+ |-l& greater j |-l.lto-10 J j-10:* ltO -2 Q ; J-4 0 .l&less J +------+ + ------+ +------+

RICHMOND RIVER. (MIDDLE) ARM WILSONS RIVER ARM 3UNOAWALBIN CREEK ARM

lEcoayatee %lle;Sxist. Dpt.D Cpt.T : Opt.A Cpt-B Opt.E \ i l e Exist.JOpt.D JOpt.F !Opt.A JOpt.B !0 p t . 1 giie;Existopt .d jopt.r ;opt.A ;opt.B facet 2 2 + SLt; 2*FI30! 4 2 ! 2*5LRJ2*FI30: * : 7 10 2*SLR!2*FI30! 4

FlOxiejren -1.2 -4.1; -6.o; -2.1; - 2 .a; -2.7; 50 0.0! 0.0! 0.0! o.o! 0.0 0.0 a : pygay -3.1 -6.7;J-16.4! -9.1! -14.7 i ^20.£ 80 0.0! Q»oi|iggj^ 0.0 ;f i. 2 2.9 -41.3 — JUJL -5.6 -6 .3 Jr W ,6! -a. a; >_g 90 -16.2,: .18.51^-33.7; -28.3! -44.6 -61.1 -5.5 -44.2 95 -19.. 3 ^ -20.3! -33.6! g P H -62.2 95 m w m - 3 3 ^ ^ 5 6 ^ ! ;estuary ...... : length -9.2; - 1 0 * 6 -48 8: -99.7 99 l& gliggG H i --*2.3i -45.3! -48.7 -*«-2 99 -44.0! -66.71 -69.3! -80.6! ■ 0 ■ 0 ■ 0 O H 0 > 30 ; -0.3 11.3 -6.2 -0.9 -1,8! -2.9. 50 0.0: 0,0 0 ■ 0 0 0 0 0 0 ! Platypus 80 ! -2.8 -4 .1 -13.0 -6.2 •11.61 -1 7 . i 80 o.o; 0,0

: <0.5 ; 9c ;j -5.6 -6 .6 -16.3 -ia .9 90 0.0; 0.0 m m -**•*: : ppc : — i------• -9 .2 -16.8 i -13.0 i p 95 -12.01-13.6 ^ 30.7J. .-25.,.3! -49.0! -65.2! 95 : -*.3 • .sP^Ssjl* _____ 1______f

; length I 99 ! -5.9 -6 .6 -15.7 -13.2 roBB: -32.*: 99 - i 7 . 7

0.0 0.0 50 0,0 0.0 0.0! 0.0 0.0 0.0: Freshwater; 20 -1.5 -3 .8 -9.7; -2.9! -4.2 J -5.3 50 0.0 ! 0.0 0.0! o.o* •cosystea !------:------0.0; 0 .0 ! 0.0 80 0.0 0.0 o.o; 0.0 0.0 0.0! *cod£catf. ; 50 ; -5-1 -7 .7 f ; - * 3 -11 : so 0.0 0.0 o.o; ♦eacrophy. ------: -- - - - t ; 90 0.0 0.0 o.o! 0.0! 0.0 90 0.0 0.0 -8.5 m m -32.7; ; 9C : w-f2 .8 -1 4 . 8 ______; <1.0 ppt .j-T-fv— -*9-2 0.0 ! 0-0 ’ -iT T r; -2 .8 ; -7 2 .0 95 -10.4 -11.7 -23,s: .*8.1 -66.0: : 93 -14. f *32.S!^2#||: -44.3: -33.2: 95 eat | ;...... 99 -ls.Tijia'o-.? -4 1 .9 ! -42.9! -81.1! -98.8 99 -11.6 -11.3 - s 9 ' 6 -93.0. area 59 : -15 M -16.2 -60.8! ‘ EBM k

0.0 0.0 0.0 30 0.0 0.0 o.o; 0.0: o.o; 0.0 50 0 .0 : 0.0 0.0; -3.3 - 10.0

80 0.0! 0.0 0.0; 0.0! 0.0 ; 0.0 - 1.8 -11.4 -5.2 -13.6 90 0,0! 0.0 0.0; 0.0! 0.0! 0.0 90 0.Q 0.0 0.0 ! 0.0 0.0 0.0; 0.0 0.0 o.o! 0.0! 0.0* 0.0 95 0.0 0,0, o.o ! 0.0 0.0 0.0; 93 J -7-3 -6.7 -X2L.3, 95

99 99 0.0 0.0, o .o ; 0.0 0.0 ',.-32.3: 59 : -10.9 -8.»S3K *mnmu +

0.0 0.0 0.0! 0.0! 0.0; 0.0! 0.0 0,0 0.0! 0.0 0.0 o.o: baas: 30 11.0; * 74.3 *14.6; *27.1! 30 50 ing 0.0 0.0 0.0! 0.0: 0.0; 0.0: $0 0,0 0.0 0.0; 0.0 0.0 0.0: so : +3.3 a.a.* * 8.4 -10.0! 42.9! 80 0.0; 0.0! 0*0! o.o; 0.0 0.0 0.0: bet 90 -2.1 -3. 7: *11.6! *4.9! *36.5! *68.3! 90 0.0 0.0 90 0.0 0.0 0.0! 5413 0.0! 0.0! 0.0: 0.0 0.0: 95 : * 15.3 ♦5.6J *44.2! *42.4; *72.95*117.5! 95 0.0 0.0 0.0! 95 0.0 0.0 0.0; 0.0

q q 0.0; 0,0; 0.0; *7,3! *29.0! *31.8! *69.3!♦111.5; 99 0.0 0-0 . ; 99 0.0 0.0 o.o; 0.0 0.0 0.0;

0.0 0.0 0.0! 0.0! 0.0: 0,0 50 0.0 0.0 0.0: 0.0 0.0 0.0: T I Sydn -3.1 ! -9.s:jpgjggj -ii.5_;r-i7v4: 50 AJoyst 0.0 0.0! 0.0! 0.0: 0.0 80 0.0 0.0 0.0; 0.0 0.0 0.0: 0.0; 0.0! 0.0 0.0; 0.0; 80 0.0

0.0 0.0 0.0! 0.0! 0*0; 0.0 90 0.0 0.0 0.0! 0.0 0.0 0.0: 0.0; 0.0! 0.0 o.O! 0.0: 90 0.0 G.O 0.0! 0.0! 0.0; 0.0 95 0.0 0.0 0.0: 0.0 0.0 0.0; 0.0: 0.0; 0.0 0.0! 0.0; 95 0.0 0.0 0.0! 0.0: 0.0! 0.0 99 0.0 0.0 0.0! 0.0 0.0 0.0; 0.0; 0.0 0.0: 0.0: 99

Unisearch SUMMARY OF PERCENTAGE CHANGE IN THE LENGTH/AREA OF Figure WRL ESTUARY WITH SALINITIES SUITABLE FOR 3.8 SIX ECOSYSTEM FACETS Report No. 99/51 99069-3-8 cdr Risk index (sum of weighted socxee)

0 2 4 8 8 10

-®- ffiohmond Wilsons BungswsiWn

Differences between the Richmond (Main), Wilsons and Bungawalbin arms are shown. The extraction rate origins are: 2 = Existing, 4 = Option A, 7 = Option B and 10 = Option E.

Unisearch RELATIONSHIPS BETWEEN THE RISK INDEX AND THE EXTRACTION Figure WRL RATE SURROGATE (M1/Ha/yr) FOR FOUR FRESHWATER-CONTENT 3.9 PERCENTILE CONDITIONS OF THE RICHMOND RIVER ESTUARY teport No. 99/51 99069-3-9.cdr A ppendix A

Preliminary Desktop Analysis of Richmond Estuary Vegetation including Information on the Salinity Tolerances of Flora

Final Draft November 1999 for K A Bishop

by

M G Dalby-Ball, G R Sainty and S Jacobs TABLE OF CONTENTS 1 Summary 3 2 Introduction 4 3 Results 6 3.1 W hat is known of the aquatic and riparian vegetation of the Richmond River estuary 6 3.1.1 Other studies outside the Richmond River catchment 6 3.1.2 Studies within the Richmond River catchment 7 3.2 Summary of the known or estimated salinity tolerances of aquatic plants and riparian vegetation likely to be in the Richmond River estuary— freshwater end 9 3.2.1 Tolerance: 9 3.2.2 Physiological effects 9 3.2.3 Limitations of experimental data 9 3.2.4 Limitations of field study data 11 3.2.5 Available information on salinity tolerances 12 3.2.6 Sources of information that will be available later this year 14 4 Discussion 16 4.1 Setting salinity‘levels’. 17 4.1.1 Issues with setting ‘acceptable’ salinity levels: 18 4.2 Future research to identify ecological thresholds in relation to salinity for aquatic plants and riparian vegetation— freshwater end 19 4.3 Conclusions 19 5 References 21 6 Annexure I. Emigrant Creek Vegetation Assessment 24

Salinity measurements are given in parts per thousand (ppt) to be consistent with the main report. Conversions from ppt to Microseimens per centimeter (pS cm'1) are given periodically.

Riparian and aquatic flora of the Richmond River Salinity tolerances Sainty and Associates November 1999 PRELIMINARY DESKTOP ANALYSIS OF RICHMOND ESTUARY VEGETATION

1 Summary Riparian vegetation and aquatic macrophytes grow along and in the Richmond River Estuary. Condition of aquatic vegetation has been classed as poor along the entire length of the river estuary, with Bungawalbyn being in the best condition (Bishop, 1999b; Appendix B). Day (1994) reports good habitat value (aquatic vegetation) for fish in Bungawalbyn Creek. Some areas are cleared to the waters edge while some areas have a band of vegetation greater than 100 m wide (Bishop, 1999b; Appendix B). Details of species presence and distribution in the Richmond River is limited to that by West et a l (1985) who mapped mangroves, saltmarsh and and the study in Emigrant Creek, a coastal tributary of the Richmond, (Sainty and Jacobs, 1999) where 41 species of native aquatic macrophyte species were recorded. Very limited field observations on mangroves and the Common Reed (Phragmites australis) (Bishop, 1999b; Appendix B) indicated that mangroves dominate the edge vegetation to ~ 35 km upstream where Phragmites australis becomes dominant.

A literature review of salinity tolerances of aquatic and riparian vegetation, revealed high variability in reported tolerance limits for some species. Salinity limits determined from experimental studies and field observations vary greatly between and within species. Tolerance limits of plants are suggested to be greater in systems where salinity fluctuates due to freshwater inputs and tides.

Information on the response of aquatic plants and riparian vegetation to salinity is too variable to make generalised predictions for the Richmond River. A range of responses is expected for different species. To maintain species diversity, salt sensitive species need to be identified and their tolerances tested under the specific conditions that occur in the Richmond River. Field observations from fresh water wetlands in South Australia suggest a general limit of 4ppt (5882 fiS cm'1) for wide ranging Australian freshwater macrophytes (Brock, 1981; Brock and Lane, 1983).

Some Australian freshwater macrophytes showed declining condition when salinity levels exceed 1 ppt (1471 ji S cm'1) (Hart et al., 1990, 1991). It is recommended that salinity levels, initially, do not exceed 1 ppt in areas that are currently fresh water and 6ppt in the , Phragmites australis transition zone (~35km upstream).

Published tolerance limits are available for a few riparian species (Marcar and Crawford, 1996) but applicability of the results to the Richmond River is uncertain. It could be expected that some plants could decline in condition if their roots are exposed to saline soil > 2ppt (2941 pS cm'1). Accordingly it is

Riparian and aquatic flora of the Richmond River Salinity tolerances Sainty and Associates November 1999 recommended that soil salinity should be below 2ppt in areas that are currently unaffected by salt.

Phragmites australis is an edge species that may die-off from the lower reaches of the river if salinity levels rise above these plants tolerance level (reported tolerance limits range from 1.5ppt-65ppt but mostly 5-25ppt (7353-36765 pS cm'1). At present this plant is occurring in a zone where the 95th percentile salinity ranges from 4.6-7.3ppt (6765-10588 pS cm'1). The other 5% of occasions salinity levels are above these levels (data supplied by Water Research Laboratory, $990). 4ke vVj AAtcV^o The proposed extraction of water could increase^salinity in this zoneKto 6.7ppt (98/53 pS cm'1) Option A or up to 9.56ppt (14,059 pS cm'1) Option E. Currently this zone has 5.9ppt as the 95 percentile this is the lower salinity tolerance limit recorded for Phragmites australis. Increasing the salinity above 6ppt may in this zone may adversely affect Phragmites australis in this zone. The current location of P hragm ites is possibly due to a number of interacting factors including the and competition. Bank steepness increases at the same location as the start of the Phragmites occupied zone (Appendix B). Steep banks are unfavorable for mangrove establishment thus P hragm ites have a place to dominate. P hragm ites has been observed at a similar tidal level in other estuaries (Howland, 1998) however data on geomorphology is unavailable.

The proposed water salinity limit of 1 ppt (1471 pS cm'1) in the currently freshwater section of the river and 2ppt (2941 pS cm'1) soil salinity is an interim measure and should remain until further information is gained regarding actual species presence, distribution and field responses to increased salinity. Salinity limits can be re-assessed once this information is available.

Extraction of fresh water should be kept to a minimum during times when salt sensitive juveniles are growing. More research is needed to determine when such times occur.

2 Introduction Distribution and abundance of plants in estuarine habitats is influenced by salinity levels and the frequency and duration of exposure to saline water (Hart et al., 1990; Zedler et al., 1990; Montague and Ley, 1993; van der Brink and van der Velde, 1993; Montague and Ley, 1993). Changing the salinity regimes in estuaries/rivers affects the distribution and abundance of submerged vegetation (Twilley and Barko, 1990) emergent aquatic vegetation (Hart et al., 1990; Hocking, 1981; Lissner and Schierup, 1997; Geddes, 1987) and riparian vegetation. Effects of salinity on plants include reduction in: (i) growth rate, (ii) flowering and leaf production and in sensitive species, death (Hart et al., 1990). Changes in species composition also occur in changed salinity conditions due to

Riparian and aquatic flora of the Richmond River 4 Salinity tolerances Sainty and Associates November 1999 some species having a competitive advantage over others under differing salinity (Kenkel e t a l., 1991). Salinity levels in the Richmond River range from saline (sea-water strength) at the estuary mouth to fresh water above the tidal limit. The salinity structure of the estuary was modeled by the Manly Research Laboratory (see main report).

Plants provide habitat for animals. Vegetation, aquatic and riparian has been associated with the presence of fish species (Bishop, 1999a; Bishop, 1999b; Growns e t a l., 1998; Pearce, 1994) and provided physical stability to waterways (Frankenberg and Tillard, 1991). Thus an understanding of how plants respond to changes in salinity regime is critical in making informed management decisions regarding changing freshwater flows.

Studies on the Richmond River estuary include extracting data from aerial photographs on the extent of riparian vegetation (Bishop, 1999a), mapping mangroves, saltmarsh and (West e t a l., 1985); mapping features (Earley, 1999) and a study on vegetation in Emigrant Creek (Sainty and Jacobs, 1999). The latter study provides information about the presence and abundance of flora species in this section of the Richmond. Channel geometry of the Richmond River including channel width, maximum and mid-channel depth, submersed bank slope, channel diversity index (based on channel width, depth and submerged slope) was determined for each 5 km of the estuary by Bishop (1999a). Bathymetry data used for these calculations was obtained from profiles recorded in the early 1980s by the NSW Department of Public Works. An assessment of stream health was conducted for the Richmond River Estuary in 1997 (Bird, 1997).

This study addresses issues relating to potential vegetation changes in response to expected increases in salinity below the tidal limit and saline incursions upstream due to extraction of fresh water from the upper estuary of the Richmond River. The following issues are addressed: • what is known of the aquatic vegetation and riparian vegetation of the Richmond River estuary. • summary of the known or estimated salinity tolerances of aquatic plants and riparian vegetation likely to be in the Richmond River estuary—freshwater end. • future research to identify ecological thresholds in relation to salinity for aquatic plants and riparian vegetation are presented.

Riparian and aquatic flora of the Richmond River Salinity tolerances Sainty and Associates November 1999 3 Results

3.1 What is known of the aquatic and riparian vegetation of the Richmond River estuary

Sea grass, saltmarsh and mangrove communities have been mapped (Summers, 1997; West e t a l, 1985). Information on the distribution, abundance and ecology of aquatic macrophytes in the Richmond River is limited to a study by Sainty and Jacobs (1999) in Emigrant Creek, a tidally influenced tributary of the Richmond, and from a report on fish sampling that noted the presence of aquatic and riparian vegetation (Pearce, 1994).

Observations from other river systems and experimental studies indicate macrophyte distribution is influenced by: salinity (Geddes, 1987; Clarke and Allaway, 1993; Lissner and Schierup, 1997; Hart e t a l., 1991; van der Brink and van der Velde, 1993 and Wijck e t a l., 1994) substrate (Clarke and Allaway, 1993), sunlight (Twilley and Barko, 1990; Clarke and Allaway, 1993) and competition (Kenkel e t a l. , 1991).

3.1.1 Other studies outside the Richmond River catchment A study on the Clarence/Coffs Harbour Regional Water Supply Project, found no information was available on aquatic macrophytes in the upper estuary of the River (Howland, 1998). Distribution of dominant aquatic and riparian habitats in the Clarence River estuary was mapped from a boat (Howland, 1998). Mangroves ( Avicennia marina and Aegiceras corniculatum ) dominated the riparian and edge vegetation to about 35 km from the river mouth. Common Reed ( Phragm ites australis) then became the dominant species and continued to be so for a further 50 km into the freshwater upper reaches of the river (Howland, 1998). Two narrow tributaries, one within 15 km from the estuary mouth, one 35 km were dominated by P h r a g m ite s (Howland, 1998). Distribution of edge/riparian vegetation in the Richmond River is expected to be similar to that in the Clarence system and inferences for the Richmond are drawn from that study. Observations in the Richmond River indicate the same distribution of mangroves and Common Reed.

An Environmental Impact Study on the Shoalhaven Water Supply Augmentation carried out in 1996 (Ecology Lab, 1996) found macrophytes grow along most, if not all, of the river within the study area. Species identification and distribution were not recorded in detail but the following observations were made: seagrasses ( Z o s te r a ) and mangroves occupy the zone from the estuary mouth to approximately 30 km upstream; P h r a g m ite s sp. becomes dominant for 10 km; Vallisneria sp. occupies the zone upstream of the P h r a g m ite s and continues to do so for about 6 km to a freshwater pool where lilies and other floating macrophytes were observed (Ecology Lab, 1999).

Riparian and aquatic flora of the Richmond River o Salinity tolerances D Sainty and Associates November 1999 Large rafts of water hyacinth were observed in the estuary near Woodburn and up to 20km upstream (Bishop pers obvs, 1999).

3.1.2 Studies within the Richmond River catchment The Richmond River Catchment Wetlands Inventory was produced by Earley (1999), this document contains maps of wetland features and is the most complete habitat map to date. Wetland features mapped are; estuarine waters, mangrove forests, dunal swamp and water bodies, upland swamp and water body, open fresh water, swamp complex and swamp (Earley, 1999). Some details of plant species found in the differing zones were recorded.

The Richmond River Catchment Stream Health Assessment Report was partially reported by Bird (1997). Riparian vegetation was assessed in terms of width and species diversity while aquatic vegetation was assessed on abundance and diversity. For details of quality ratings (see Bird, 1997 or Appendix B). Results were summarised by Bird (1997) on two spatial scales, regional (4 regions) and sub-catchments (4-7 sub-catchments per region). Ratings covered five categories from very poor to very good.

Conclusions of findings for riparian vegetation, at the regional level, ranged from the highest to lowest grades (‘very good’ to ‘very poor’). Aquatic vegetation was classed as ‘poor’ to ‘very poor’. Bungawalbyn Creek had the highest ratings for health in terms of riparian and aquatic vegetation. Ratings for each section of the Richmond River estuary are given below.

Location Riparian vegetation Aquatic macrophytes Upper 65% poor 98% very poor Richmond Arm 35% moderate 2% poor

Wilson’s River 10% very good 100% very poor Arm 10% good 30% moderate 2% poor 30% very poor

Bungawalbin 50% very good 85% very poor Creek Arm 30% moderate 15% poor 20% ven/ poor

Lower 20% good 100% very poor Richmond Arm 38% moderate 42% very poor

Riparian and aquatic flora of the Richmond River ~J Salinity tolerances Sainty and Associates November 1999 NB: This data is specifically for the estuary. Results from sub-sections were not available at the time of writing this report.

Brackish and Freshwater Macrophytes A study of the vegetation in Emigrant Creek, a tidally influenced tributary of the Richmond, by Sainty and Jacobs (1999) located the following vegetation:

Submerged species: 6 native and 1 exotic Floating species: 2 native, 2 exotic Floating-attached species: 2 native, 1 exotic Emergent species 31 native Riparian species: 21 native, 29 weed

Species are given in Annexure I

Salinity appeared to have some influence on presence/abundance of emergent and submergent macrophytes in Emigrant Creek. General trends being a decrease in abundance (or presence) of salt sensitive species further downstream, for example Persicaria hydropiper, and P. s tr ig o s a and an increase in the presence/abundance of salt tolerant species further downstream, examples being Common Reed, Phragmites australis, and Mangrove Fern, Acrostichum speciosum (Annexure I). Some species only grew in the upper, freshwater reaches of Emigrant Creek (e.g. Eleocharis sphacelata and Is a c h n e g lo b o s a ) (Sainty and Jacobs, 1999).

Distribution and abundance of some species in respect to salinity was different to expectations based on experimental data and field observations. Examples of anomolies include the presence of the submerged plants, Potamogeton ja v a n ic u s and Ceratophyllum demersum —species thought to have medium salt tolerance, in the lower reaches of the creek where the salinity can be high. The emergent macrophytes, Persicaria hydropiper and P. s tr ig o s a—moderately salt tolerant, extend further downstream than expected (Annexure I) (Sainty and Jacobs, 1999).

Riparian Riparian flora has been disturbed/removed by land clearing over the past century. Clearing has, and continues to, affect the integrity of the riparian vegetation communities. A greater number of weed species than (29 weed and 21 native species Annexure I) were recorded in the riparian vegetation along Emigrant Creek (Sainty and Jacobs, 1999).

Integrity of riparian vegetation along the Richmond River Estuary (all arms) was calculated from aerial photographs (Bishop, 1999a, Appendix B). Riparian canopy along the Richmond River estuary was measured in 1 km intervals from coastal Ballina inland 115 km to Casino. For details of the aerial photograph study refer to Appendix B (Bishop, 1999a). A transition zone from mangroves

Riparian and aquatic flora of the Richmond River Salinity tolerances 8 Sainty and Associates November 1999 (.Avicennia marina and Aegiceras corniculatum) to Common Reed ( P h r a g m ite s a u s tr a lis ) was noted between 32 and 38 km from the estuary mouth between the towns of Broadwater and Woodburn.

3.2 Summary of the known or estimated salinity tolerances of aquatic plants and riparian vegetation likely to be in the Richmond River estuary—freshwater end

Plant communities have been identified by Hart e t a i , (1990, 1991) as the wetland biota most sensitive to increases in salinity. Most research on the effects of salinity on plants has focused on physiological responses of agricultural species (Greenway and Munns, 1980; Yeo e t a i , 1988). Information about responses of Australian Macrophytes to salinity is predominantly form limited field observations and some experiments (Brock, 1981; Brock and Lane 1983a and 1983b; James and Hart, 1993; Hart e t a i., 1990; 1991; Warwick and Bailey, 1997). James and Hart (1993) and Hart e t a/.,(1991) concluded that 4ppt (5882 pS cm'1) appears to be the upper tolerance limit for widespread freshwater macrophytes. A common finding is that species with low or no tolerance to salinity have no, or reduced mechanisms, for excluding ions particularly Na+ (Montague and Ley, 1993; Twilley and Barko, 1990).

3.2.1 Tolerance: Salinity tolerance of a plant is where the plant is able to maintain (regulate) the optimum internal osmotic gradient. In this report tolerance is defined as: the salinity level at which basic function of a plant becomes impaired to the extent that its long-term survival, in its current location, is unlikely. Where a tolerance limits is given from other sources the definition of tolerance is that associated with the particular study. Tolerance levels of individual plants make up the general tolerance limit of that species, however care should be taken when ascribing tolerance limits to species based on recorded levels for individuals as genotypes of single species often show great plasticity in response to salinity.

3.2.2 Physiological effects Where the salinity level is greater than a plant’s tolerance limit the plant may not be able to keep its cells turgid due to either a lack of water and/or an excess of ions. Saline conditions can be lethal to freshwater plants because the cells can not maintain turgor (Hart e t a i , 1991). Sub-lethal effects on macrophytes include reduced growth rate, size and flowering, death of juveniles and inability to uptake water leading to water stress.

3.2.3 Limitations of experimental data Salinity is not an easy variable to manipulate in the field hence studies on plant tolerance to salinity are usually conducted in ‘controlled conditions’. Limitations of studies attempting to determine salinity tolerance by experimentation include:

Riparian and aquatic flora of the Richmond River 9 Salinity tolerances Sainty and Associates November 1999 • Salinity in controlled experiments is almost always a constant level whereas in field salinity levels fluctuate. Fluctuations in salinity levels may be as important as mean salinity in determining salinity tolerance and distribution of macrophytes (Montague and Ley, 1993; Warwick and Bailey, 1997). Montague and Ley (1993) suggested that standard deviation of salinity rather than mean salinity is a better predictor of macrophyte distribution and abundance than mean salinity.

• Duration of submergence (contact) with saline waters also influences a plants response to salinity (Hart e t a l., 1990). Field observations suggest some species will tolerate higher than average salinitities where levels fluctuate, as in estuaries. An example of this is the presence of Schoenoplectus vallidus (a fresh water species) growing between Phragmities australis and mangroves where it is occasionally inundated by moderately saline water (G. Sainty pers obvs, 1999). Thus tolerance limits for freshwater species derived from species in wetlands (such as those suggested by Brock, 1981; Hart e t a l, 1991, 1994) may underestimate the ability of plants in estuarine habitats to withstand increased salinity levels.

• Controlled experiments traditionally examine the responses of mature plants however different lifecycle stages often have different tolerances to salinity. Germination and establishment usually require optimal environmental conditions. Germinating seeds and juvenile plants are likely to be less tolerant, than adults, to increases, or greater fluctuations, in salinity, examples include Juncus kraussii, Typha orientalis (Zedler e t a l., 1990) and Phragm ites australis (Lissner and Schierup, 1997).

• Studies are often of short duration (about 8 weeks) and do not follow through generations. Plant response to salinity may be gradual with long-term growth being effected. Short-term experiments, commonly 8 weeks, may not detect long term effects of salinity changes.

• Sexual reproductive capabilities of plants can be effected by salinity. Reduced flowering was recorded in Potamogeton perfoliatus (Twilley and Barko, 1990; van der Brink and van der Velde, 1993), Myriophyllum spicatum (Twilley and Barko, 1990) and two species of non-Australian Potamogeton (van der Brink and van der Velde, 1993) exposed to salinity levels above their usual range.

• Frequently species are tested as monocultures, where as in field situations competitive interactions between species are likely to influence plants response to salinity. Studies that have addressed the issue of interspecific competition in relation to salinity changes found species showed large differences in their salinity response when grown as a mixed culture compared with their response as a monoculture (Kenkel e t a l., 1991; Zedler e t a l., 1990). Some species that grew well at high salinity levels as a

Riparian and aquatic flora of the Richmond River -t Salinity tolerances ' u Sainty and Associates November 1999 monoculture were suppressed at all but the lowest salinity, by competitive species, when grown as a mixed culture (Kenkel e ta i, 1991).

• Plants collected for experiments are invariable from one location and results are often generalised for the species. Such interpretation can lead to incorrect assumptions about salinity tolerance of a species as salinity tolerance has been shown to vary on a local and geographical scale with plants collected from areas of higher salinity showing greater tolerance that those areas of lower salinity (Lissner and Schierup, 1997). Examples of salinity tolerance for Phragmites australis range from 5-65ppt depending on where the study plants originated with most reports indicating a tolerance limit between 5-25ppt (7353-36765 pS cm'1) (Clucas and Ladiges, 1980; Lissner and Schierup, 1997). Genotypes of the saltmarsh plant Spergulina marina were found to differ in their phenotypic responses to changes in salinity (Redbo-Torstensson, 1994). Considerable variability in salt tolerance has been found between individuals of the same species (Marcar and Crawford, 1996) hence the sample size of plants in experiments and records of variability are very important as small sample size and mean values will not provided accurate information.

• Conclusions drawn from experimentally obtained tolerance limits need to remain open when making decisions about how plants will respond in the field. External conditions can interact to influence a plants response including a lack of water can lower a plants resistance to salinity (Lissner and Schierup, 1997) , as can increased nutrient levels.

3.2.4 Limitations of field study data

Aquatic vegetation In the field the following confounding confounding factors influence the observed salinity tolerance limits of aquatic vegetation: • Topography, substrate and light • Competition • Ground water and flushing with freshwater following rain • Nutrient status

Riparian vegetation Most information regarding salinity tolerance of trees has been obtained from monitoring growth and survivorship of trees grown on salt affected land. Results from such studies are often site specific due to the confounding effects from physical components of the environment, particularly the soil-water and nutrient status and biological components such as mycorrhizae, pathogens, and competitors. Considerable variability in salt tolerance has been found

Riparian and aquatic flora of the Richmond River -j -| Salinity tolerances Sainty and Associates November 1999 between individuals of the same species (Marcar and Crawford, 1996) further reducing the reliability of stated tolerance limits for such species.

3.2.5 Available information on salinity tolerances Reported ‘general’ tolerances A review of the salt sensitivity of Australian freshwater biota concluded that salinity levels of around 1 ppt (1471 pS cm'1) appear to be the threshold for adverse biological effects to occur in Australian fresh water systems (Hart e t a i , 1991). Information forming the basis of this conclusion came from experimental studies and field observations. Ecological field observations such as that of Brock (1981) found many fresh water species do not tolerate salinity greater than 1 —2ppt (1471-2206 |iS cm'1) and that by 4ppt (5882 pS cm'1) most freshwater macrophytes are absent. Caution is necessary here as not all freshwater plants are salt sensitive; plus this assumes complete soil and water saturation and this does not occur in estuaries.

The work of Bailey e t a i , has been extended to form a large study, funded by the Water Resources Corporation, looking at the response of Australian biota to salinity. This document will be published later this year and should be consulted prior to making final determination of the appropriate mean salinity level and level of fluctuation that should be achieved.

Emergent vegetation Information on abundance and diversity of aquatic macrophyte species, in the Richmond River, is limited to a study by Sainty and Jacobs (1999) on Emigrant Creek. Field and experimental studies from wetland plants indicate freshwater macrophytes die off when salinity levels exceed ~2ppt species are absent where salinity exceeds 4ppt (Brock, 1981; Brock and Lane 1983; Hart e t a i , 1991). As stated previously, tolerance limits may be greater in areas where saline periods are interspersed by freshwater periods or where plant roots have access to fresh ground water.

Experimental tolerance limits for emergent macrophytes that may occur in the Richmond River include:

Eleocharis acuta upper limit or 2ppt, with reduced stem height high and non­ replacement of stems above this salinity level (Hart e t a i , 1990).

Typha domingensis was found to be moderately salt tolerant with reduced growth at 50 mM NaCI (4297 pS cm'1).

Phragmites australis will survive 25ppt (36,765 pS cm’1) for one or two months (Harris as cited in Bishop, 1995). Severe damage occurs to this species if exposed to sea strength salinity (35ppt; 51, 471 pS cm'1) for a few days (Sainty and Jacobs, 1981). Field observations of Phragmites australis showed die-back occurred when water salinity increased (value not stated Clucas and Ladiges

Riparian and aquatic flora of the Richmond River -| p Salinity tolerances Sainty and Associates November 1999 1980) and when soil salinity was greater then 15ppt (Lissner and Schierup, 1997). Experimental work with this species indicated low mortality <15% at 12ppt (17,647 pS cm 1) and all plants died at 35ppt (51, 471 pS cm'1) salinity (Lissner and Schierup, 1997).

Schoenoplectus vallidus (recorded in Emigrant Creek) is generally a freshwater species but has been observed in growing between Phragmites australis and mangroves where it is periodically inundated by moderately saline water of about 15ppt (22,059 pS cm'1) (Sainty pers obvs, 1999).

Field observations of moderately salt-tolerant macrophytes extending into areas that are expected to have salinity levels above that expected for the species may be explained by the presence of fresh ground water around plant roots or by the dynamics of saltwater inundation including the duration and frequency of exposure to saline water. Edge vegetation may have reduced exposure to saline water as mixing of fresh and saline water takes time and salt water usually moves upstream at depth in a wedge (O’Loughlin pers comm).

Submerged vegetation There is a paucity of data on salinity tolerance of submerged macrophytes but available information indicates a range of salinity tolerance limits.

Experiments conducted to determine the salinity tolerance of Potamogeton perfoliatus (a species recorded in Australian river systems) provide conflicting evidence with suggested tolerance limits of <1.5ppt (2,206 pS cm'1) (Metcalf, 1931) while (1932) measured peak growth at 4.2ppt (6,176 pS cm'1) and a tolerance of 11.2ppt (16,471 pS cm'1). Field observations in the US suggest P. perfoliatus can survive being periodically flushed with 10—12ppt salinity water (Twilley e t a l. 1985). Tolerance of up to 12ppt (17,647 pS cm'1) was recorded by Twilley and Barko (1990).

Potamogeton pectinatus (recorded in Emigrant Creek), has been recorded growing in water with a salinity level of 3.6ppt to 6.3ppt in Lakes in Victoria (Sainty e t a l., 1990). Yezdani (1970) records this species as having a high tolerance and growing where salinity reached 10ppt (14,706 pS cm'1).

Vallisneria americana, Ribbon Grass. Vallisneria nana was recorded in Emigrant Creek by Sainty and Jacobs (1999) this species was previously incorrectly named \/. a m e r ic a n a (Jacobs pers comm) thus results for V. americana are given in this report for V. nana). Vallisneria americana was considered a freshwater species (Metcalf, 1931) and said to thrive at 1.5ppt (Sainty and Jacobs, 1981) but has been observed growing in an estuary with a mean salinity of 5.8ppt, range of Oppt to 12.8ppt (Davis and Brinson, 1976). Twilley and Barko (1990) recorded unaffected growth of V. americana in a controlled experiment at salinities up to 12ppt (17,647 pS cm'1). Vallisneria americana is resident near

Riparian and aquatic flora of the Richmond River Salinity tolerances Sainty and Associates November 1999 the mouth of the and is reported to repeatedly colonise the area after dilution from high salinity (Sainty pers comm, 1999).

Hydrilla verticillata, a species recorded from Emigrant Creek (Sainty and Jacobs, 1990) is reported to have low tolerance to salinity with field observations in areas of < 2ppt salinity (Rybicki e t a l, 1985) and experimental results of little above 4ppt (Twilley and Barko, 1990). One record contradictory to other results is growth up to 13ppt (19118 pS cm'1) salinity (Steward and Van, 1987).

Riparian Most trees and shrubs are not salt tolerant and experience reduced growth even at low salt concentrations (Marcar and Crawford 1996). Salt may be excluded or excreted from plants. Trees generally have salt exclusion mechanisms in their roots. Plants adapted to saline conditions may have salt excretion glands in leaves (e.g. mangroves) or divert salt to old leaves (Yeo, 1994).

Five of the 21 species of native riparian plants recorded from Emigrant Creek (Sainty and Jacobs, 1999) have been classed according to their salinity tolerance Marcar and Crawford (1996), details of how results were obtained were not given in the study. Marcar’s study classes trees into four groups from slight salinity tolerance (2-4 dSrrf1) to extreme (>16 dSm'1)

Tolerance Species Slight tolerance (2-4 dSm'1) Acacia melanoxylon Eucalyptus grand is

Moderate tolerance (4-8 dSm'1) Cassurina cunninghamiana Melaleuca quinquenervia

Severe tolerance (8-16 dSm'1) Casurina glauca

Extreme (>16 dSm'1) Avicennia marina Aegiceras corniculatum

A data-base of ‘species occurrence in relation to salinity, is currently being completed by Dr Paul Bailey (pers comm. Dr Bailey, Monash University) and should be consulted when it becomes available later this year.

3.2.6 Sources of information that will be available later this year • Data Base— Salt sensitivity data-base of Australian fresh water biota. A data­ base of the distribution of Australian biota in relation to salinity levels is currently being completed by Dr P. Bailey, Monash University. This work is

Riparian and aquatic flora of the Richmond River Salinity tolerances Sainty and Associates November 1999 expected to be available later this year (Produced with funding from the Land and Water Corporation).

• Review and Recommendations for Salinity—National Dry Land Salinity Program (Bailey; Hart e t a l 1999, Report for the Land and Water Corporation). This report documents the sensitivity of Australian Fresh Water biota to salinity including aspects of management and policy. To be published later this year.

Riparian and aquatic flora of the Richmond River ■] 5 Salinity tolerances Sainty and Associates November 1999 4 Discussion Few studies have published information on the presence and abundance of flora species in the Richmond River, particularly downstream of the tidal limit.

Dominant Edge Vegetation Mangroves and Common Reed dominate much of the water edge with a transition zone from Common Reed to mangroves about 35 km from the estuary mouth (Ecology Lab, 1996). This is similar to the Shoalhaven and Hastings Rivers (Howland, 1998; Ecology Lab, 1996). Both mangroves Avicennia marina (Clarke and Allaway, 1993) and Common Reed Phragmites australis (Lissner and Schierup, 1997; pers obvs) grow in fresh water.

Phragm ites australis was noted to die-back when water salinity increased (value not stated Clucas and Ladiges 1980) and when soil salinity was greater then 15ppt (Lissner and Schierup, 1997). Experimental work with this species indicated all plants died at 35ppt salinity (51,471 pS cm'1) (Lissner and Schierup, 1997). Hence Phragmites australis would be limited in its downstream distribution by salinity.

Results imply that salinity levels will influence the downstream extent of P h r a g m ite s but not the upstream extent of mangroves. Salinity values in the transition zone of the Richmond River are between 7.2ppt and 4.6ppt (95% of the time) this suggests that the current downstream limit of P h r a g m ite s is unlikely to be due to salinity alone and is more likely to be influenced by competition from mangroves. One studied reported 5ppt as being a tolerance limit for this species as so it can not be said that salinity is not the controlling factor in the Richmond River estuary. P h r a g m ite s may have a competitive advantage over mangroves in areas where bank slope is greater than 2/10 m and this may explain the transition zone in an area where bank slope increases above 2/10 m. Distribution of mangroves initially depends on distribution of propagules (Clarke and Allaway, 1993) but where the shore line has a steep gradient the potential does not exist for for of , this lack of sediment precludes the establishment of mangroves or allows only a fringe to develop (Hartly, 1997).

If abstraction of freshwater from the upper Richmond River results in salinity levels in the transition zone exceeding the tolerance limit for Phragm ites australis these plants will die-back as they did in the Gipsland Lakes (Clucas and Ladiges, 1980). If mangroves are inhibited from colonising this area by bank slope in excess of 2/10 m the area may be colonised by a species other than P h r a g m ite s or mangroves, or remain unvegetated. Changes in species composition may be gradual and alternate between species in relation to environmental conditions with P h r a g m ite s dominating the banks during prolonged freshwater flows.

Riparian and aquatic flora of the Richmond River Salinity tolerances Sainty and Associates November 1999 Emergent and submerged vegetation The inability of plant cells to maintain turgor in water with elevated salinity results in reduced vigor and may be lethal. Wide ranging results of salinity tolerances between studies, observations and experiments, high light the limitations of applying stated tolerance levels to plants in a particular location. Field and experimental studies indicate freshwater macrophytes experience detrimental effects when salinity levels exceed ~2ppt and most freshwater species are absent where salinity levels exceed 4ppt (Hart etal., 1991).

Information on abundance and diversity of aquatic macrophyte species, in the Richmond River (Sainty and Jacobs,1999) suggest the presence of emergent and submergent species that cover a range of salinity tolerances. If freshwater is extracted above the tidal limit it can be expected that saline water will reach areas that are now fresh. Emergent and submerged freshwater species may die-out if salinity levels exceed plants tolerance limits. Total abundance may not change but species composition probably will. Insufficient data is available to make predictions on changes in abundance of particular species.

Emigrant Creek—vegetation and salinity Salinity readings were not taken during this study and direct links between vegetation and salinity are not made. Saline influence is hard to judge with tidal rise and fall noted to be strong at below the tidal limit in a part of the stream that is likely to be fresh water. Results from the study of vegetation in Emigrant Creek indicate presence/abundance of emergent and submergent macrophytes in the creek may be influenced by salinity levels. Some species only grew in the upper, freshwater reaches, of Emigrant Creek (e.g. Eleocharis sphacelata and Isachne globosa) and others only in the saline areas Avicennia marina. General trends were a decrease in abundance (to the point of being absent) of moderately salt tolerant species further downstream, for example P e r s ic a r ia hydropiper, and P. strigosa and an increase in the abundance of salt tolerant species further downstream, examples being Common Reed, P h r a g m ite s a u s tr a lis , and Mangrove Fern, Acrostichum speciosum (Annexure I). (Sainty and Jacobs, 1999).

The occurrence of species with moderate salinity tolerance in downstream, saline areas, may be due to either high rainfall resulting in a flushing effect with freshwater continually pushing the saline water downstream or the groundwater may be fresh providing a predominately fresh or only slightly saline substrate for the plants root systems. A further alternative is that these species have a higher tolerance to salinity than previously reported.

Community composition in backwaters Freshwater backwaters are usually floristically diverse and the vegetation classification (Bird, 1997) suggest many backwaters exist along the Richmond River. Those below the point of proposed increased freshwater extraction may experience an overall increase in dryness (if they are away from tidal influence)

Riparian and aquatic flora of the Richmond River -| y Salinity tolerances Sainty and Associates November 1999 or increased salinity (if they occur within tidal influence). A reduction in species diversity may occur under either circumstance.

Riparian Results of tree species tolerance to salt is not complete enough to make predictions regarding the effect of increased salinity below the tidal limit of the Richmond River. Evidence suggests that most riparian species have low to moderate salt tolerance. Some tree death and/or reduced growth may occur where salt concentration in the root zone increases above the tolerance limit of the plants. Specific predictions can not be made for individual species. Negative effects of increased salinity are likely to be amplified if the vegetation is lacking in freshwater (i.e. if there is a drought) or there is a nutrient imbalance. A suggested salinity level for soil in the root zone of riparian vegetation currently occuring along the freshwater section of the Richmond River is <2ppt, this limit is based upon information from Marcar and Crawford, 1996 on salinity tolerance of a few riparian plants. This salinity level may be reassessed after 1 year of observation of riparian vegetation in the salt-effect zone. It is likely that increases in soil salinity will result in the spread of Cassuarina glauca at the expense of other native species.

4.1 Setting salinity ‘levels’. It is inevitable that salinity levels will rise as a result of fresh water extraction and the potential effects of this on the current submerged and emergent plant communities ranges from no change to plant death and/or a change in species composition. Given this and taking into account and the paucity of data for the species in question, the ‘acceptable’ salinity limit should be conservative and open to assessment following results from monitoring after low-level water extraction. Further measures should be taken ensure the ability to modify salinity by strategic freshwater releases.

4.1.1 Issues with setting ‘acceptable’ salinity levels: Salinity levels fluctuate with the tide, rainfall events and other variables. It is possible that set ‘acceptable’ salinity levels will be exceeded during periods when low rain coincides with high tides thus increasing the relative salinity of the water.

Reduction in fresh water flow due to low rainfall is very likely to coincide with greater demand for fresh water again resulting in salinity levels increasing above mean levels.

As discussed, plant response to salinity is often dependent on plant age and reproductive state with juvenile and reproductive plants being more sensitive to increased salinity than adults. Knowing this, extraction of fresh water should be kept to a minimum during times when salt sensitive juveniles are growing. More research is needed to determine when such times occur.

Riparian and aquatic flora of the Richmond River 18 Salinity tolerances Sainty and Associates November 1999 4.2 Future research to identify ecological thresholds in relation to salinity for aquatic plants and riparian vegetation—freshwater end • Mapping and classification of riparian and aquatic vegetation communities associated with the Richmond River.

• Update with current research. Dr Kimberly James and post-graduate research students are currently working on understanding the relationships between salinity and the distribution/abundance/tolerance of aquatic macrophytes.

• Long term, before and after, replicated field experiment and observational studies of aquatic and riparian vegetation in east-coast river systems that are expected to undergo (new or increased) freshwater extraction (see Underwood, 1998 for appropriate experimental design). Estuary systems could be the Hastings, the Shoalhaven and the Clarence.

• Sampling in geomorphologically similar areas that differ in salinity (Montague and Ley, 1993).

• Experiments—being aware of the limitations of experimental studies it is suggested that further studies could include: • field transplant experiments where species, with their substrate intact (Grillas e t a l., 1992) are transplanted to an area of differing salinity. Such studies will inherently have confounding variables such as differing substrate type, period of inundation, strength of flow etc. In designing the experiment, variables should be noted and attempts should be made to execute the experiment such that the influence of confounding variables are kept to a minium including the use of appropriate controls. Examples of such experiments are given in Underwood (1998).

• long-term experiments where plants are grown in differing salinity from juvenile to adult and through reproductive periods.

4.3 Conclusions Information pertaining to the distribution and abundance of riparian and aquatic vegetation in the Richmond River is limited. Salinity tolerance limits for aquatic vegetation is available for a few species although their applicability to specific areas is unknown. Distribution of Australian freshwater macrophytes appears to be limited to areas where salinity is < 4ppt, with diminished plant productivity where salinity > 2ppt Plants with some tolerance to salinity such as P h r a g m ite s a u s tr a lis may experience a contraction of its seaward extent if salinity levels exceed the tolerance limit of those plants growing in the Richmond estuary. Mangroves may extend further upstream if P. australis retreats and if P. australis had previously competed with the mangroves, however this will not occur if bank steepness procludes mangrove establishment.

Riparian and aquatic flora of the Richmond River Salinity tolerances Sainty and Associates November 1999 If soil salinity exceeds the tolerance limits of riparian vegetation, reduced growth and possible plant death will occur. However, it is not possible to give the salinity level this would occur at, due to the paucity of data and the confounding effects of the site.

It is recommended that salinity levels in areas that are now freshwater should not exceed 1 ppt and soil salinity levels along freshwater sections of the river do not exceed 2ppt. Salinity in the transition zone of mangroves and P h r a g m ite s a u s tr a lis (~36km upstream) should not exceed 6ppt. Salinity levels can be reassessed following further field work, including experimental work and observations of changes in aquatic and riparian vegetation after extraction of freshwater.

Different of different species can have very different salinity tolerances. ‘Salinity tolerance’ is too simplistic an approach to determine environmentally aware extraction quotas of freshwater. A more meaningful approach is to think holistically about the saline environment. Field-work is necessary to address this question and work is required both in the Richmond estuary and in other ‘reference’ rivers and estuaries in NSW.

Riparian and aquatic flora of the Richmond River Salinity tolerances 20 Sainty and Associates November 1999 5 References Bird, L., (1997). The Richmond River Catchment Stream Health Report. An ecological assessment of the condition of Coastal Streams in the Richmond River Region of Northern NSW. Unpublished Report to DLWC, Nov. 1997.

Bishop, K. A. (1995). Hastings Dristrict water supply augmentation EIS; Aquatic ecology study for the Intake pump station upgrading at Koree Island. Unpublished report for Connell Wagner.

Bishop, K. A. (1999a). Revised eight-part test on 3 threatened freshwater fish species in relation to the Pacific Highway upgrading: The Ballina Bypass. Unpublished Report to Connell Wagner.

Bishop, K. A. (1999b). Appendix B. Characterisation of fish communities and habitat within the upper arms of the Richmond River estuary, with emphasis on the potential effects of altered salinity structure as arising from freshwater extraction in the upper estuary. Study undertaken for the Water Research Laboratory on behalf of the NSW Department of Land and Water Conservation.

Brock, M. A. (1981). “The ecology of halophytes in the south-east of South Australia.” Hydrobiologia, 81, 23-32.

Brock, M. A., and Lane, J. A. K. (1983). “The aquatic macrophyte flora of saline wetlands in Western Australia in relation to salinity and permanence.” Hydrobiologiaogia 105, 63-76.

Brock, M. A., and Shiel, R. J. (1983). “The composition of aquatic communities in saline wetlands in Western Australia.” Hydrobiologia, 105, 77-84.

Clark, P.J. and Allaway, W.G. (1993). The regeneration niche of the Grey Mangrove (A v ic e n n ia m a rin a): effects of salinity, light and sediment factors on establishment, growth and survival in the field. Oecologica, 93: 548-556.

Clucas, R. D., and Ladiges, P. Y. (1980). Die-back of Phragmites australis (Common Reed) and increased salinity in the Gippsland Lakes.” 292, Ministry of Conservation, Melbourne, Victoria.

Ecology Lab (1996). Shoalhaven Water Supply Augmention EIS. Aquatic Ecology and Fisheries Downstream of Burrier. Unpublished report to Dames and Moore.

Frankenberg J. and Tillard J. (1991). Protecting Riverbanks from . Australian Planner. June 1991. Pp 107-110.

Geddes, M.C. (1987). Changes in salinity and in the distribution of macrophytes, and fish in the Coorong , SA, following a period of River Murray Flow. Trans. R. Soc. S. Aust. 111(4): 173-181.

Grillas, P., Van-Wijck, C., and Boy, V. (1992). Transferring sediment containing intact seed banks: a method for studying plant community ecology. Hydrobiologiaogia 228: 29-36.

Growns e t al. (1998) Fish Hart, B. T., Bailey, P., Edwards, R., Hortle, K., James, K., McMahon, A., Meredith, C., and Swadling, K. (1990) . “Effects of salinity on river, stream and wetland ecosystems in Victoria, Australia.” Wat. Res., 24(9), 1103-1117.

Hart, B. T., Bailey, P., Edwards, R., Hortle, K., James, K., McMahon, A., Meredith, C., and Swadling, K„ (1991) . “A review of the salt sensitivity of the Australian freshwater biota. Hydrobiologiaogia 210: 105-

Riparian and aquatic flora of the Richmond River Salinity tolerances Sainty and Associates November 1999 144.

Harty, C. (1997). Mangroves of NSW and Victoria. Pulsa Vista Publications. Hocking, P. J. (1981). “Response of Typha domingensis to salinity and high levels of manganese in the rooting medium.” Aust. J. Mar. Freshwat. Res., 32, 907-919.

Howland, B. M. (1998). Clarence/Coffs Harbour Regional Water Supply Project. Environmental Assessment Working Paper No. 9. Estuarine Flora and Fauna. July 1998. Centre for Coastal Management Southern Cross University P.O. Box 5125 East Lismore NSW 2480. Unpublished report to DPWS NSW. Pp34

James, K. R., and Hart, B. T. H. (1993). “Effect of salinity on four freshwater macrophytes.” Australian Journal of Marine and Freshwater Research, 44, 769-777.

Kenkel, N.C. Mcllraith, A.L. Burchill, C.A. and Jones. G. (1991). Competition of three plant species to a salinity gradient. Can. J. Bot. 69: 2497-2502.

Lissner, J. and Schierup, H. (1997). Effects of salinity on the growth of Phragmites australis. Aquatic Botany 55: 247-260.

Marcar, N. and Crawford, D. (1996). Tree-growing strategies for the productive use of saline land. Aust. J. of Soil and Water Conservation. 9 (3) 34-39.

Montague, C.L. and Ley, J.A. (1993). A possible effect of salinity fluctuation on Abundance of Benthic Vegetation and Associated Fauna in Northeastern Florida Bay. Estuaries 16 (4): 703-717.

Pearce, A. (1994). Testing the relative selectivity of various methods applicable for surveying fish in freshwater streams. Unpublished data, intergrated project for Bachelor of Applied Science (Fisheries Management and Aquaculture), Southern Cross University.

Redbo-Torstensson, P., (1994). Variation in plastic response to a salinity gradient within a population of halophytic plant Spergularia marina. OIKOS 70: 349-358.

Sainty G.R. and Jacobs S.W.L. (1999). Vegetation in Emigrant Creek. Unpublished Report to Connell Wagner.

Summers D. (1997). Review of the decline of seagrass habitats in the lower Richmond River using remote sensing techniques. Unpublished data, intergrated project for Bachelor of Applied Science (Fisheries and Aquaculture), Southern Cross University 38pp.

Twilley, R. R., and Barko, J. W. (1990). “The growth of submersed macrophytes under experimental salinity and light conditions.” ESTUARIES, 13(3), 311-321. van der Brink and van der Velde (1993) van der Moezel, C. S., Walton, C. S., Pearce-Pinto, G. V. N., and Bell, D. T. (1988). “The response of six E u c a ly p tus species and Casuarina obesa to the combined effect of salinity and waterlogging.” Australian Journal of Plant Physiology, 15:465-474. van der Moezel, P. G., and Bell, D. T. (1987). “The effect of salinity on the germination of some Western Australian E u c a ly p tus and M e la le u ca species”, 15, 239-246. van Wijck, C., Grillas, P., de Groot, C. J., and Ham, L. T. (1994). “A comparison between the production of Potamogeton pectinatus L. and Myriophyllum spicatum L. in the Camargue (southern France) in relation to salinity and sediment characteristics.” Vegetatio, 171-180.

Riparian and aquatic flora of the Richmond River Salinity tolerances 2 2 Sainty and Associates November 1999 Yedzani, G.H. (1970). A study of the quarternary vegetation history in the volcanic lakes region of western Victoria. Ph.D. thesis, Monash Univ., Melbourne (as cited in Aston, H.I., (1973). Aquatic Plants of Australia. Melbourne University Press. Pp368

Warwick, N. W. M., and Bailey, P. C. E. (1997). “The effect of increasing salinity on the growth and ion content of three non-halophytic wetland macrophytes.” Aquatic Botany, 58, 73-88.

West (1995). An estuarine inventory for NSW, Australia. Zedler, J. B., Paling, E. and McComb, A (1990). Differential responses to salinity help explain the replacement of native Juncus kraussii by Typha orientalis in Western Australian . Aust. J. of Ecol. 15: 57-72.

Riparian and aquatic flora of the Richmond River 23 Salinity tolerances Sainty and Associates November 1999 6 Annexure I. Emigrant Creek Vegetation Assessment Prepared by Surrey Jacobs and Geoff Sainty Date of inspection 16-17/4/99

Species presence and abundance from each of the seven locations surveyed along Emigrant Creek; common (+++), widespread or patchy, not common (++) present but infrequent (+). ? = Species identification not confirmed or species currently under taxonomic review.

Sites are located from the upper tidal limit being mostly fresh water (section 1) to sea strength (section 7) with section 3 having a salinity level fluctuating around 2/3rds sea strength and sections > 4 being mostly sea strength (pers comm G. Sainty, 1999). Sandy flat creek is usually freash water.

SUBMERGED

Site location 1 2 3 4 5 6 7 Sandy Flat Ck *Cabomba caroliniana1 Ceratophyllum demersum + + ++ H ydrilla verticil lata + ++ +++ +++ ++ + ++ Lim nophila ?aromatica +++ +++ ++ _|—1_ + Potam ogeton javanicus + + +++ +++ Potam ogeton pectinatus ++ ++ Vallisneria nana + + + ++ ++ 1 Not recorded in the reaches inspected however there is a large in festation in Emigrant Creek . FLOATING Emiq rant Cree<—U Dper Estuary Site location 1 2 3 4 5 6 7 Sandy Flat Ck Azolla pinnata * + + ++ +++ ++ + Lem na/Spirodeia 1 *Salvinia m olesta ++ +

FLOATING-ATTACHED Emigrant Creek—Uoper zstuary Site location 1 2 3 4 5 6 7 Sandy Flat Ck Ludwigia peploides + ++ ++ *Nymphaea caerulea + ++ +++ +++ + +++

Riparian and aquatic flora of the Richmond River 24 Salinity tolerances Sainty and Associates November 1999 Nymphoides indica ++ ++ ++ +++ ++ +

EMERGENT Ernierant Creek—UpperEstuary Site location 1 2 3 4 5 6 7 Sandy Flat Ck Acrostichum speciosum + + + ++ +++ +++ +++ Cladium procerum + Commelina cyanea +++ +++ +++ +++ +++ + Crinum pedunculatum + ++ ++ ++ ++ ++ +++ C y a th e a ?a u s tr a l is ++ + Cyperus ?gunnii + + + + Cyperus polytachyos ++ C y p e r u s sp. C y p e r u s sp. + Eclipta prostrata + + + + + Eleocharis sphacelata + + Eleocharis ?equisetina/dulcis ++ Hibiscus diversifolius + + + + + + + H y p o le p is sp. + Isachne globosa + + Juncus polyanthemus + + Leersia hexandra ++ ++ +++ + ++ + ++ Lomandra hystrix ++ + + + Ludwigia octovalvis + + + + + + Oplismenus ?aemulus ++ Paspalum distichum ++ + + + Persicaria attenuata + Persicaria decipiens + + + + + Persicaria hydropiper +++ +++ +++ ++ + ++ Persicaria strigosa +++ +++ +++ +++ ++ Philydrum lanuginosum + + Phragm ites australis ++ +++ +++ +++ +++ +++ Schoenoplectus mucronatus ++ + + Schoenoplectus validus + + + Triglochin procerum +++ +++ +++ +++ +++ ++ + Typha domingensis +++ + ++ +++ +++

Riparian and aquatic flora of the Richmond River 2 5 Salinity tolerances Sainty and Associates November 1999 RIPARIAN Ernierant Creek—Upper_Estuary Site location 1 2 3 4 5 6 7 Sandy Flat Ck

Acacia melanoxylon + + A c a c ia sp. + Acmena smithii + + Aeqiceras corniculatum + + ++ Avicennia marina ++ H—I—h Casuarina cunninqhamiana ++ ++ Casuarina qlauca ++ ++ Cupaniopsis anacardiodes + + D ry maria cord at a ++ + Eucalyptus q rand is + Ficus coronata + + + ++ + F ic u s sp. + + + Flaqellaria indica + Hibiscus tiliaceus ++ H—f" +++ ++ M adura cochinchinensis + M elanthera biflora + + Melaleuca quinquenervia ++ + ++ ++ ++ + ++ Pittosporum undulatum + Sambucus ?gaudichaudiana + Schefflera actinohylla + W aterhousea floribunda ++ ++

EMERGENT RIPARIAN Ernierant Creek—Uaper Estuary Site location 1 2 3 4 5 6 7 Sandy Flat Ck WEEDS *Aqeratina adenophora + *Ageratina riparia ++ ++ *Ageratum conyzoides + + + + *Anredera cordifolia ++ * Aster subulatus + ++ + + *Axonopus affinis + *Baccharis halim ifolia + ++ + *Urochloa mutica +++ +++ +++ +++ +++ +++ +++ +++ *Buddieja davidii +++ * C a n n a sp. ++ ++ *Co!ocasia esculenta ++ +++ ++ + *Cinnamomum camphora +++ +++ ++ ++ ++ +++

Riparian and aquatic flora of the Richmond River Salinity tolerances 26 Sainty and Associates November 1999 *Dipogon lignosus + *Echinochloa crus-galli + + + + ++ *Echinochloa pyram idalis + *Erythrina crista-galli + + ++ ++ *Erythrina sykesii +++ +++ +++ +++ +++ + *lpomoea alba ++ + + *lpomoea cairica +++ +++ +++ +++ +++ +++ +++ *Lantana camara +++ +++ ++ +++ ++ ++ *Ligustrum lucidum ++ *Macroptilium atropurpureum ++ *Myriophyllum aguaticum ++ ++ ++ ++ + + +++ *Paspalum wettsteinii + *Ricinus communis + *Sagittaria platyphylla ++ ++ *Salix babylonica + *Senecio mikanioides ++ *Solanum mauritianum + + + + +

Riparian and aquatic flora of the Richmond River 27 Salinity tolerances Sainty and Associates November 1999 A ppendix B

Characterisation of Fish Communities and Habitats within the Upper Arms of the Richmond River Estuary with emphasis on the Potential Effects of Altered Salinity Structure arising from Freshwater Extraction in the Upper Estuary

(Preliminary Desktop Analysis)

25 November 1999

by

Dr K A Bishop Consultant SUMMARY

* The present investigation was commissioned by the Water Research Laboratory on behalf of the NSW Department of Land and Water Conservation (DLWC). The study arises from increasing pressures to extract freshwaters from the upper arms of the Richmond River estuary. To manage these extractions in an ecologically-sustainable manner, the DLWC is seeking a sound basis for assessing extraction-vs-environmental tradeoffs. The central focus of this process are impact mechanisms involving shifts in the salinity structure of the estuary, particularly in the upper estuary.

* The fish communities and fish habitats of the upper arms of the Richmond River estuary were the focus of the present desktop investigation. The primary objective of the work was to gather and analyse pertinent information necessary for assessing the potential impacts of predicted shifts in the estuary's salinity structure. This involved: a) the characterisation of fish communities and their habitats, b) the identification of high- value components, particularly those most vulnerable to shifts in salinity structure, and c) the identification of salinity- exposure characteristics which best typify transition conditions where freshwater fauna are displaced by more salt- tolerant fauna, or where freshwater habitat is degraded.

* There were clear indications that the Bungawalbin Creek arm of the estuary potentially contains extensive high-value physical habitat. Unfortunately, there are also indications that the value of this habitat has recently been reduced by deteriorating water quality. To obtain more definition and confidence in the identification of high-value habitats, habitat surveys along the upper arms of the estuary would need to be undertaken.

* The upper Richmond River arm of the estuary potentially has the greatest complement of high-value fauna given the additional records of the high-conservation-value Oxleyan pygmy perch and Eastern freshwater cod. These records need to be confirmed and this would only be possible through intensive, well-focussed sampling. *

* There were many inconsistencies amongst and between indicative salinity tolerances derived from the literature/discussions and those derived from the analysis of available data along salinity gradients. These inconsistencies reflect the complexities involved in understanding the way in which salinity influences fish communities in the upper arms of estuaries. Given an understanding that the majority of Australian freshwater fish have a relatively short evolutionary history in freshwaters, there is likely to be a reduced importance of direct physiological impacts arising from increases in salinity. Instead more attention should generally be focussed on i) direct impacts causing adverse behavioural changes, and ii) indirect impacts such as habitat degradation or competition and predation arising from species with markedly-high salinity tolerances.

3 * Five indicative salinity limits arose from the study. Given that they are based on 'best-available' information, they can be used as 'working' thresholds to be inputed into risk analyses which assess the implications of extracting particular volumes of freshwater. The risk-analysis process helps to identify characteristics of extractions (e.g. volume, rate, and timing) which should minimise resultant impacts. The process as described has the advantage that it attempts to utilise what information is available, or readily attainable, on individual estuaries and their most valuable and/or vulnerable components. Of fundamental importance, the process utilises estuary-specific information on known links between hydrology and basic structural properties of the ecosystem.

* Given the complexities of estuarine ecosystems and inaccuracies which may occur in the methodologies, particularly in relation to the range of working thresholds utilised, it is imperative that any implemented extraction regime be viewed as an interim condition, to be revised once substantial knowledge is gained through ensuing scientific research and monitoring. This is a fundamentally important feature of any adaptive management system.

* Monitoring in estuarine ecosystems is potentially more difficult than in river/stream ecosystems due to the complexities introduced by the daily tidal cycle. To enable comparisons between places and times, samples should be taken at the same phase of the tidal cycle, and the same time of day. Attempts to utilise comparable combinations of tidal phase, and time of day, will result in protracted sampling sessions.

h TABLE OF CONTENTS

odd 1

all 1.0 INTRODUCTION...... 5 1. 1 Background...... 5 1.2 Present focus & other facets...... 5 1.3 Why were fish considered?...... 5 1.4 Objectives of the present investigation...... 6 1.5 Impacts of freshwater extraction on estuaries...... 6 1.6 Influence of salinity on estuarine fish communities . 8 1.7 Zone classification in estuaries...... 9 1.8 Terminology in relation to salinity tolerance..... 10

2.0 METHODS...... U 2.1 Characterisation of fish communities & their habitats....11 2.2 Identification of high-value and vulnerable components...13 2.3 Identifying salinity-exposures likely to cause change....14

3.0 RESULTS...... 16 3.1 Characterisation of fish communities & their habitats.... 16 3.1.1 Fish habitats...... 16 3.1.2 Fish fauna...... 21 3.2 Identification of high-value and vulnerable components... 26 3.2.1 Fish habitats...... 26 3.2.2 Fish fauna...... 27 3.3 Identifying salinity-exposures likely to cause change....28 3.3.1 Literature/discussion derived salinity tolerances...... 28 3.3.2 Analysis of data gathered along salinity gradients...... 31

4.0 DISCUSSION AND CONCLUSIONS...... 38 4.1 High value components...... 38 4.2 Inconsistencies between indicative salinity tolerances... 38 4.3 Using indicative salinity limits...... 39 4.4 Adaptive management and monitoring...... 40

5.0 ACKNOWLEDGMENTS...... 41

6.0 REFERENCES...... 42

WITHIN TEXT:

TABLES 1-7

FIGURES 1-24

5 1.0 INTRODUCTION

1.1 Background

The present investigation was commissioned by the Water Research Laboratory on behalf of the NSW Department of Land and Water Conservation (DLWC). The study arises from increasing pressures to extract freshwaters from the upper arms of the Richmond River estuary. To manage these extractions in an ecologically- sustainable manner, the DLWC is seeking a sound basis for assessing extraction-vs-environmental tradeoffs. The central focus of this process are impact mechanisms involving shifts in the salinity structure of the estuary, particularly in the upper estuary (MHL 1998). A detailed background of the study, including maps of the estuary, is given in the Main Report.

1.2 Present focus and other facets of the broader investigation

The fish communities and fish habitats of the upper arms of the Richmond River estuary are the focus of the present, pilot desktop investigation .

Estuarine ecosystems have a vast number of biotic (living) and abiotic (non-living) components and linkages. It is therefore never possible to make meaningful predictions concerning ecosystem health by considering just one facet of the ecosystem. For this reason a range of ecosystem facets were targeted in the broader investigation. That is the broader investigation takes a multifaceted approach which attempts to account for the potentially high level of ecosystem complexity. In this context, two parallel investigations, targeting other components of the estuarine ecosystem, accompany the present investigation: *

* Estuarine plants (Dalby-Ball, Sainty & Jacobs): Appendix A * Platypus (Grant): Appendix C

1.3 Why were fish considered?

Fish were included for a wide range of reasons:

Direct susceptibility:

* some species are directly susceptible to changes in salinity through physiological stress, * some species are directly susceptible to changes in salinity through adverse behavioural changes (avoidance, loss of migration and/or navigation cues, etc)

Indirect susceptibilty:

* some species are indirectly susceptible to changes in salinity through increased competition and predation from salt-tolerant species * some species are indirectly susceptible to changes in salinity through salinity-mediated impacts on their food organisms * some species are indirectly susceptible to changes in salinity through the loss of habitat arising from salinity mediated impacts on plants for example

6 High-value features:

* fish have an important ecological role as primary consumers (detritivores) and secondary & tertiary consumers (carnivores),

* fish have traditionally been used in assessments (Hellawell 1986)

* fish have a high public-appreciation, recreational and commercial value (Bunn e t a l . (1998) indicated that there are several reasons for choosing particular 'target' or indicator species: the inclusion of fish species of commercial or economic importance provides an opportunity to estimate the potential economic impacts of flow diversion)

* some taxa have a high conservation value (notably Eastern freshwater cod, Maccullochella ikea, and Oxleyan pygmy perch, Nannoperca oxleyana , which occur in the Richmond system, and are afforded legal protecting under the Fisheries Management Amendment Act 1997)

1.4 Objectives of the present investigation

The primary objective of the work was to gather and analyse pertinent information necessary for assessing the potential impacts of predicted shifts in the estuary's salinity structure. This involved:

* the characterisation of fish communities and their habitats, particularly in the upper estuary

* the identification of high-value components, particularly those most vulnerable to shifts in salinity structure, and *

* the identification of salinity-exposure characteristics which best typify transition conditions where freshwater fauna are displaced by more salt-tolerant fauna, or where freshwater habitat is degraded (emphasis being placed on understanding the limitations of the thresholds obtained and the complications involved in their use)

1.5, Impacts of freshwater extraction on estuaries

Drinkwater and Frank (1994) presented one of the most comprehensive reviews of the effects of freshwater regulation and diversion on the adult and larval stages of fish and invertebrates in coastal marine waters including estuaries. The authors described declines in coastal fisheries noting their general association with reductions in freshwater inflow. These were described in relation to effects on migration, spawning success, advection of eggs and larvae, species competition and distribution, general productivity, food supply, and water quality. It was emphasised that extensive ecological considerations are required during the planning stage of freshwater-modification projects to minimise potential impacts.

7 Within Australia, Glaister (1978) and Ruello (1973) described reductions in commercial fish catches in NSW estuaries in relation to natural decreases in freshwater inflows. Bunn et a l . (1998) summarised a range of similar incidences in .

Bishop (1999a) assessed the potential impacts of large-scale water diversions on the fisheries within the Clarence River estuary and the adjacent marine environment. An initial stage of this work involved the identification of major mechanisms by which impacts may occur (a large proportion of the identified mechanisms arose from the review of Drinkwater and Frank [1994]). Twelve major mechanisms were identified and these are listed in Table 1: one relates to impacts on flow variability, seven concern impacts on low-end flows, and four relate to middle and high-end flow impacts. A subsequently identified mechanism concerning the upstream navigation and transport of fish and invertebrates into and along estuaries is also included in this table. This mechanism arises from Odum (1970).

Given the potential-scale of freshwater extractions from the upper arms of the Richmond estuary, only the mechanisms concerning impacts on low-end flows have relevance to the present study. Even within this group, there are only four mechanisms which are likely to particular relevance:

* FRAIE3- - reduced low-end inflows resulting in extended durations of elevated salinities in the upper-middle estuary; fauna with low salinity tolerance (eggs, larvae, juveniles or adults) will be adversely affected through physiological stress and/or by competition and predation from colonising large fauna normally found in the lower estuary; Odum (1970) indicates that the low-salinity region of an estuary acts as a nursery ground important for the protection of juvenile fish and invertebrates

* FRAIE4 - reduced low-end inflows resulting in extended durations of elevated salinities in the upper-middle estuary; plants with low salinity tolerance will be adversely affected through physiological stress

* FRAIE5 - reduced low-end inflows resulting in extended durations of elevated salinities in the lower estuary; marine biota thus able to colonise the lower portion of the estuary *

* FRAIE13 -reduced low, middle or high-end inflows which subsequently dissipate salinity & other chemical gradients out from the mouth of the estuary, and/or along the lower-middle estuary; this is significant as there is evidence (Odum 1970) that many juvenile estuarine fish & invertebrates use such gradients to navigate there way into and along estuaries. Salinity-gradient upstream transport mechanisms could also be inhibited. Reductions in abundance, biomass and diversity could result.

8 TABLE 1 CHECKLIST OF MAJOR MECHANISMS BY WHICH LARGE-SCALE WATER DIVERSIONS MAY LEAD TO IMPACTS ON THE FISHERIES WITHIN ESTUARIES AND ADJACENT MARINE ENVIRONMENTS. From Bishop (1999a), initially done in relation to the Clarence River estuary. A subsequently identified mechanism concerning the upstream navigation and transport of fish and invertebrates is included at the end of this list (FRAIE13).

* FRAIE1 - altered variability of inflows to the estuary, and the consequent change in patterns of variation in the salinity structure of the estuary, is likely to disrupt life cycles as suitably-timed breeding and/or migration cues for fish and are masked; growth/ opportunities are lost because of a lack of synchronization with the temperature regime. Reductions in abundance, biomass and diversity of the fauna results.

* FRAIE2 - reduced low-end inflows resulting in hostile water- quality conditions (e.g. low DO at depth) in deep sections within the upper-middle estuary where water retention times are protracted; demersal eggs and large-size taxa are at most risk because they are found in deeper sections where water quality is likely to be most hostile; reductions in egg survival, abundance and growth will occur

* FRAIE3 - reduced low-end inflows resulting in extended durations of elevated salinities in the upper-middle estuary; fauna with low salinity tolerance (eggs, larvae, juveniles or adults) will be adversely affected through physiological stress and/or by competition and predation from colonising large fauna normally found in the lower estuary; Odum (1970) indicated that the low-salinity region of an estuary acts as a nursery ground important for the protection of juvenile fish and invertebrates

* FRAIEA - reduced low-end inflows resulting in extended durations of elevated salinities in the upper-middle estuary; plants with low salinity tolerance will be adversely affected through physiological stress

* FRAIE5 - reduced low-end inflows resulting in extended durations of elevated salinities in the lower estuary; marine biota thus able to colonise the lower portion of the estuary

* FRAIE6 reduced low-end inflows resulting in extended durations when flow-induced currents can not suspend eggs or larvae in the upper-middle estuary; eggs or larvae settle to the bottom and mortality results

cont./- TABLE 1 cont.

* FRAIE7 - reduced low-end inflows resulting in extended durations when flow-induced currents can not transport eggs or larvae in the upper-middle estuary to favourable habitats for later life-history stages (inhibition of advection); growth/recruitment opportunities are lost and reductions in abundance, biomass and diversity of the fauna results.

* FRAIE8 - reduced low-end inflows aggravating pollution problems in the upper-middle estuary originating from agricultural pollution sources; lowered dilution of pollutants and/or stratification-induced deoxygenation causing the releases of toxicants from estuary-bed sediments; consequent lowered abundance of fish, shellfish and Crustacea, and contamination of tissues

* FRAIE9 - reduced middle and high-end inflows greatly altering the frequency that the bed of the upper-middle estuary is flushed of fine sediments and organic material (i.e. high flows causing substrate turnover); this is significant as many fish species lay their eggs on or within hard substrates - the presence of sediment/organic matter will result in lowered reproductive success as suitable egg deposition/attachment sites will become limited

* FRAIE10 - reduced middle and high-end inflows greatly altering the frequency that organic material deposited on the bed of deep sections in the upper-middle estuary is flushed out; this is Significant as a high organic load can result in hostile water-quality conditions; again demersal eggs and large-sized taxa are at most risk; reductions in abundance and growth will occur

* FRAIE11 - reduced middle and high-end inflows greatly reducing channel-maintenance processes (mediated by flushing flows) in the upper-middle estuary with a result that major habitat contraction occurs in the longterm; deep sections of the estuary are most vulnerable as very large flows are required to remove infilling material; again demersal eggs and large-sized taxa are at most risk

* FRAIE12 - reduced middle and high-end inflows subsequently reducing the input of natural river-borne nutrients and organic material; reduced thence abundance along the length of the estuary and into adjacent coastal areas; fish and abundance reduces in response to lowered food supply

cont./- TABLE 1 cont.

Subsequent ly identified mechanism:

* FRAIE13 - reduced low, middle or high-end inflows which subsequent ly dissipate salinity & other chemical gradients out from the mouth of the estuary, and/or along the lower-middle estuary; this is significant as there is evidence (Odum 1970) that many juvenile estuarine fish & invertebrates species use such gradients to navigate there way into and along estuaries. Salinity-gradient upstream transport mechanisms could also be inhibited. Reductions in abundance, biomass and diversity could result. 1.6 The influence of salinity on estuarine fish communities

The structure of fish assemblages in estuaries has been shown to be influenced by a range of environmental variables:

* salinity: (Weinstein e t al. [1980], Allen [1982], Potter e t a l. [1986], Loneragan e t a l. [1986], Marshall and Elliot [1998], Peterson and Ross [1991], Whitfield [1996], Sheaves [1998], Theil et al. [1995], Cyrus and Blaber [1987a,b], Marias [1988], Pollard and Hannan [1994]), * water temperature: (Hoff and Ibara [1977], Quinn [1980], Allen [1982], Loneragan et a l. [1986], Peterson and Ross [1991], Marshall and Elliot [1998]) * turbidity: (Blaber and Blaber [1980], Peterson and Ross [1991], Whitfield [1996], Blaber [1997], Cyrus and Blaber [1987a,b], Marias [1988]) * current speed: (Blaber 1997) * dissolved : (Blaber and Blaber [1980]) * distance from the estuary mouth: (Quinn [1980], Loneragan et al. [1986], Grey et a l. [1990], Pollard and Hannan [1994]) * sediment characteristics including heterogeneity: (Marchand [1993], Blaber and Blaber [1980]) * aquatic-plants: (Marchand [1993], Blaber and Blaber [1980]) * other influencing variables: water depth, bank slope, coarse cover provided by submersed logs and fine cover provided by submersed leaflitter, twigs and root material.

The identification of the above influencing variables has primarily arisen from investigations focusing on estuary-long gradients in environmental conditions. Up until the recent work of NSW Fisheries in the Richmond and Clarence River estuaries (originally reported in West [1993a&b], and later in West and Walford [1999]), no such investigations had been undertaken in large estuaries in eastern Australia. NSW Fisheries' work involved sampling along the main deepwater channels of the Richmond and Clarence estuaries at 17 and 12 longitudinal sites respectively. Replicated, quarterly samples were taken by trawling over a two-year period. The number species collected decreased with increasing distance from the mouth of the estuary in both rivers.. Based on the composition of catches, three significantly-different groups of sites were identified and these were viewed to correspond to the marine, brackish and freshwater sections of the estuaries. However, West and Walford (1999) concluded that it remained difficult to determine which environmental variables were most responsible for structuring these fish communities. This was because many of the variables were confounded (e.g. increased turbidity being associated with lowered salinity), and many important variables remained unmeasured.

When assessing the role of salinity, it is clearly important to recognise the complexities introduced by confounding environmental variables. It is also important realise that the tolerance of fish to extremes of this variable differs

9 considerably between species (Kinne [196A], Whitfield e t al. [1981]), particularly those in deep waters compared with those in shallow waters (Loneragan e t al. [1987]), and between different size classes of the same species (Kinne [196A], Holliday [1971]).

TEL (1996) stated that the salinity regime of an estuary is a fundamental determinant of the distribution of much its flora and fauna. This may be true, however, a large array of complex observations are required to isolate out the influence of salinity from the considerable range of other influencing environmental variables.

1.7 Zone classification in estuaries

Rochford's classification

The theory that some degree of longitudinal zonation exists within estuaries was advanced and explored by Rochford (1951). He described four zones in estuaries based on hydrological and chemical (water and sediment) data:

* Marine: - oceanic waters (negligible diurnal changes/conflict in salinity) - deep water, small tidal area - mainly marine sands

* Tidal: - less oceanic influence (slight diurnal changes/conflict in salinity) - shallow waters with extensive intertidal - scoured channels with black mud deposits

* Gradient: - waters mainly influenced by freshwater (diurnal changes/conflict in salinity at a maximum) - water confined to narrow channels - variable substrates: black muds and coarse sands

* Freshwater: - waters virtually always freshwater - water confined to narrow channels with in shallows - substrates varying from coarse sands to small boulders

No biological data were available for Rochford to evaluate the ecological significance of these zones. Many biological estuarine-gradient studies have since been undertaken, yet none have examined the ecological reality of Rochford's classification in any significant level of detail. As noted above, West (1993a&b), and later West and Walford (1999), superficially examined longitudinal zonation in the Richmond and Clarence Rivers by examining the fish species composition of trawl catches. Sections of the estuaries were identified which appeared to correspond to the above-mentioned marine, brackish and freshwater zones.

10 Traditional classification based on salinity reviewed

TEL (1996) indicated that traditionally estuaries have been divided into five zones based on differences in salinity:

* limnetic: 0 to 0.5 ppt salinity * oligohaline:: 0 ,. 5 to 5 ppt * mesohaline: 5 to 18 ppt * polyhaline: 18 to 30 ppt * euhaline: >30 ppt

Bulger e t a l. ( 1993) reviewed this classification in the light of results from a multivariate analysis of the distribution of fish and in estuaries within the United States. It was indicated that the traditional zonation should be modified to contain an overlap of salinities:

* limnetic: 0 to 4 ppt salinity * oligohaline 2 to 14 ppt * mesohaline: 11 to 18 ppt * polyhaline: 16 to 27 ppt * euhaline: >24 ppt

TEL (1996) indicated that it was apparent that no specific assemblages of flora and fauna have been defined in relation to salinity gradients in NSW estuaries. However, some pertinent information was presented by Hart e t a l. (1991) and ANZECC (1992) in that it was indicated that salinities in excess of 1 ppt may harm freshwater communities.

1.8 Terminology in relation to salinity tolerances

Biota capable of withstanding wide variations in salinity are termed 'euryhaline'. Biota only capable of withstanding a narrow range of salinities are termed 'stenohaline'.

Biota only capable of withstanding a narrow range of high salinities are referred to as ' stenohaline-marine' species. Similarly, biota only capable of withstanding a narrow range of very low salinities are referred to as *stenohaline-freshwater' species.

The stenohaline-freshwater species are of central interest in the present investigation as, with extraction of freshwater from the upper arms of the Richmond estuary, the volume of their habitat will reduce (be compressed), with corresponding advantage being passed onto euryhaline and stenohaline-marine species.

11 2.0 METHODS

The general approach used in the investigation followed the principles employed by Bishop (1995) in an ecological assessment of the impacts of freshwater extraction on the estuarine biota of the . In that investigation key, high value biota were identified which were susceptible to salinity changes in the estuary. Pertinent literature and information from experts were then examined to determine likely critical thresholds in salinity which may limit the distribution of the biota. This information was then fed into a process which assessed impacts arising from proposed water extraction. For the present investigation, the assessment process is documented in the Main Report.

2.1 The characterisation of fish communities and their habitats

Information on the character of fish habitats and fish communities was initially obtained from pertinent literature and through interviews with those familar with the system. Contact was made (on 30/7/99) with Dr B. Eyre of Southern Cross University on the understanding that two PhD students from his faculty were currently working on estuarine-gradient, ecological issues in the Richmond River estuary. I requested information arising from these studies. The request was denied.

Fish habitats

Directly-usable information on habitats was found to be quite scarce. However, three data sources potentially containing information on habitat characteristics were located. These data sources were subsequently examined and usable information were produced in the context of estuary-long changes in conditions:

Channel geometry. The data source was a large set of cross- sectional depth profiles taken along the Richmond River estuary, including the Wilsons River and Bungawalbin Creek arms. These bathymetry data were gathered by the NSW Public Works Department in the early 1980s in the course of flood mitigation investigations. To characterise key features of fish habitat the following information was extracted from the profiles * and summarised statistically (mean and standard deviation; standardised to AHD = 0) for each 5 km of the estuary where sufficient data were available:

* channel width (water edge to water edge) * maximum channel depth (water surface to lowest point on bed) * mid-channel depth (water depth in the middle of the channel) * submersed bank slope (water depth 10m from each bank, two values from each profile) * channel structure diversity index (sum of the co-efficients of variation of the following three variables: channel width, maximum depth and submersed bank slope; cv as proportion) * the number of distinct bed anomalies (see Figure 1 for an example of these anomalies; data only presented in terms of their frequency per 5 km of estuary)

12 FIGURE 1 EXAMPLES OF DISTINCT RIVER-BED ANOMALIES (ARROWED). The cross-sectional bathymetry data was gathered by the NSW Public Works Department in the Wilsons River upper estuary (approximately 93 km upstream of the Richmond River estuary mouth). The horizontal dashed line is AHD = 0. The water level at the time of surveying is 'W.L.’. The numbers below the transect are vertical (top) and horizontal (bottom) distances in metres.

CHAINAGE 28950 SECT NOo 267 DATE OF SURVEY 07/07/82 Stream-health assessment data (partially__reported--in___Bird [19971 ) . Bird ( 1997 ) reported an ecological health assessment of the condition of coastal streams in the Richmond River Region. The assessment used a rapid field survey methodology called 'The Riverine Habitat Audit Procedure' (RHAP) developed by Dr J. Anderson of Southern Cross University, Lismore, NSW.

The survey method focused on a snap-shot approach using rapid surveys of key attributes in the stream itself, along the banks, in the , and in the lands bordering the river and stream subsections. At each site surveyed the following key attributes were recorded and condition ratings ascribed: reach environs (level of in bordering lands), channel h a b ita t (the proportions, dimensions and number of channel habitats), channel form and dimensions (the diversity of measured widths and depths), and bank sediments (sediment size diversity), stream bank condition (bank stability), bed and condition/fish passage (proportion of aggrading beds forming bars; bed stability), riparian vegetation (riparian width and diversity of vegetation), aquatic vegetation (abundance and diversity), aquatic habitat (diversity and abundance of sheltering habitats and substrates), and scenic and recreational conservation values (subjective assessment based on appearance, remoteness and regulations and structures). An overall condition rating was determined by combining ratings for each of the abovementiond attributes.

Bird (1997) summarised the results of the assessment at two spatial scales: regional level (i.e. upper Richmond River, Wilsons River, lower Richmond River, and Bungawalbin Creek), and subcatchment level within regions (4 to 7 subcatchments per region). In all regions there were a total of 798 subsections (104 to 257 per region, or 4 to 58 per subcatchment), 481 of which were surveyed.

To isolate out condition ratings for the Richmond River estuary (including major arms), it was necessary to access the subsection-level data. This was done by obtaining DLWC (draft) summary maps showing quality ratings for each subsection surveyed. These were obtained for the following key attributes:

* riparian vegetation * aquatic habitat * aquatic vegetation * overall condition

The following quality ratings were colour coded on the maps:

* very poor * poor * moderate * good * very good

To allow a frequency analysis of these ratings, the Richmond River estuary was divided up into four components. The number of

13 surveyed subsections is shown below for each component:

* upper Richmond River estuary (Wilsons River confluence to the tidal limit near Casino) = 5 subsections

* Wilsons River estuary (Richmond River confluence to the tidal limit upstream of Lismore) = 9 subsections

* Bungawalbin Creek estuary (Richmond River confluence to the tidal limit) = 4 subsections

* lower Richmond River estuary (estuary mouth to the Wilsons River confluence) = 6 subsections

The frequency analysis included an adjustment to account for variation in the length of the river encompassed by the subsections. That is, the subsequent frequency data is in the form of the percentage length of the estuary, as separated out for the various quality ratings.

Aerial photography: information on riparian vegetation. Information on riparian vegetation was sought as the integrity of fish communities is frequently found to be positively related to the integrity of the riparian vegetation (e.g. Growns e t al. [1998], Bishop [1999c]). The following sets of standard colour aerial photgraphs were the data source: Ballina 1:25,000 (1997), Woodburn 1:25,000 (1993) and Lismore (1997). For each one kilometre of the estuary (all arms), the length of the banks lined with the following riparian-canopy width categories was recorded:

* canopy less than 25 m wide ('trace' category) * canopy greater than 25 m wide * canopy greater than 50 m wide * canopy greater than 100 m wide

An indication (albeit not definitive) of the integrity of the riparian vegetation is provided through the examination different canopy widths. Two kilometres is the maximum length of bank for a single channel with uniform edges. Edge convolutions increase this maximum. A channel with an island included would have a maximum length of 4 km if uniform edges are present.

Fish fauna

In fo rm atio n on the fis h fauna p re s e n t in the system hab i-fcato was obtained from pertinent literature, database searches and through interviews with those familar with the system.

2.2 The identification of high-value and vulnerable components

Vulnerable components were taken to those which appear to be primarily restricted to the freshwater zone of the estuary. West and Walford (1999) identified the freshwater zone to be upstream of a point approximately midway between Woodburn and Coraki, i.e. approximately 50 km by the estuary upstream from the estuary mouth. Some consideration was given to more-downstream

14 components, notably the Sydney rock oysters, the culture of which may be impacted by increases in biofouling as caused by increases in salinity.

Fish habitats

Information on high value habitat components was primarily derived from the three data sources which contained information on habitat characteristics.

Channel geometry. High-value features were taken to be best represented by the following variables

* channel structure diversity index, and * the number of distinct bed anomalies

Both of these variables were considered to be surrogates for habitat diversity. A higher habitat diversity is expected to be associated with a greater diversity of fish species and enhanced sheltering opportunities.

Stream-health assessment data. Estuary arms with higher quality ratings (i.e. ’very good’ or 'good') were considered to have greater value than those with lower ratings (i.e. ’poor' or 'very poor').

Aerial photography: information on riparian vegetation. High value sections of the estuary arms were taken to be those with greater lengths of banks lined with riparian vegetation. The wider canopy-width categories were considered to be the best indicators as they are likely to have riparian vegetation with the greatest integrity (i.e. multi-layered with primarily native species).

Fish fauna

The identification of species with high commercial and recreational value primarily relied on such species identified by West and Walford (1999).

Species with high conservation value were identified by examining pertinent literature. This included the list of fish species classed as either 'endangered' or 'threatened' in the Fisheries Management Amendment Act (1997).

2.3 Identifying salinity-exposures likely to cause change

Initially, pertinent literature and discussions with specialists were used to identify any available information on salinity tolerances of the high-value and other potentially-affected species.

Analysis of data gathered along salinity gradients

It was expected that critical salinity thresholds to be obtained through the literature and discussions would be quite limited. To partially offset this problem, available estuary-long data sets were investigated, and where possible, examined in relation to parallel sets of salinity data. Of interest were sections of

15 relationships which show major changes in species abundances with small changes in salinity. It was expected that reasonably reliable salinity-exposure thresholds could be identified from such features.

Notable estuary-long datasets from the Richmond River estuary which were investigated included:

* data on fish communities from the upper Emigrant Creek estuary (Bishop 1999c) * data on fish communities from the Newrybar flood mitigation channel (Graham 1989) * data on fish communities in deep sub-tidal habitats in the Richmond River estuary mainchannel (NSW Fisheries data; partially presented in West [1993a], West [1993b] and West and Walford [1999])

Notable estuary-long datasets from other estuaries which were investigated included:

data on fish communities in deep sub-tidal habitats in the Clarence River estuary (NSW Fisheries data; partially presented in West [1993a], West [1993b] and West and Walford [1999] ) data on fish communities in shallow habitats in the Clarence River estuary (NSW Fisheries data; partially presented in West and King [1996], West [1993a] and West [1993b]) data on fish communities in tributaries of the lower Clarence River estuary (NSW Fisheries data; partially presented in Pollard and Hannan [1994]) data on littoral fish communities in the Hawkesbury-Nepean River estuary (NSW Fisheries data; partially presented in Pollard e t al. [1994] and Gehrke and Harris [1996])

16 3.0 RESULTS

3.1 The characterisation of fish communities and their habitats

3.1.1 Fish habitats

Existing broad classification

Earley (1999) completed a wetlands inventory of the Richmond River catchment providing baseline information on the extent and distribution of wetland types within the catchment. Within the classification scheme, 'estuarine waters' referred to waters influenced by tidal action and having measurable freshwater dilution from the surrounding catchment.

The 1:80,000 maps included in the inventory indicated that estuarine waters occurred along the Richmond River system only as far upstream as 61.5 km by estuary upstream from the estuary mouth (i.e. ~1.5 km downstream of the Wilsons River confluence; near Coraki). Beyond this point, including the Wilsons River arm, the estuary was classified as a "creek/river". Similarly, from the confluence point with the Richmond River, both Bungawalbin Creek and Emigrant Creek were classed as "creek/river".

These "creek/river" classifications are clearly incorrect as the tidal limits (as indicated by MHL [1998]) are well upstream of Earley's classification-change-over positions. Additionally, Bishop (1999b&c) provided data indicating the penetration of saline water within Emigrant Creek to a point 20 km upstream from its confluence with the Richmond River estuary, the point where Emigrant Creek was classified by Earley (1999) as a "creek/river" rather than "estuarine waters".

General description of the estuary.

A summary of the structure of the major arms of the Richmond River estuary is given below in relation to distances from the estuary mouth and major landmarks:

* Richmond River (middle arm):

- tidal limit (TL) ~111 km by estuary upstream from the estuary mouth following MHL (1998), or TL “113 km upstream following A. Moore pers. comm. (immediately below Grays Falls near the eastern limit of Casino)

- 'landmark' points are:

+ Ballina: “1 to 6 km upstream + Wardell: “18 km upstream + Broadwater: ~ 26 km upstream + Woodburn: “42 km upstream + Coraki: “62 km upstream + Tatham: “87 km upstream

17 * Wilsons River arm:

- commences "63 km by estuary upstream from the estuary mouth (near Coraki) - tidal limit "112 km by estuary upstream from the estuary mouth following MHL (1998), this is "12 km by by estuary upstream of Lismore. - therefore "59 km in length

* Bungawalbin Creek arm:

- commences "55 km by estuary upstream from the estuary mouth (7 km downstream of Coraki) - tidal limit "91 km by estuary upstream from the estuary mouth following MHL (1998), - therefore "36 km in length

* Emigrant Creek arm:

- commences "6 km by estuary upstream from the estuary mouth (western limit of Ballina) - tidal limit (TL) "29 km by estuary upstream from the estuary mouth following Bishop (1999b&c), - therefore "23 km in length

The Richmond estuary characteristically has sinuous river channels surrounded by broad . In its lower reaches it spilts into separate arms forming islands at several locations. Some large cutoff bays are also present in these reaches. West (1993b) indicated that its structure has been influenced by rising sea levels ('drowned river valley' character), and by marine sand and fluvial sediment infilling ('infilled barrier estuary' character).

Available information for different components of the estuary

Lower Richmond River. This 63km-long component of the estuary extends from the estuary mouth to the Wilsons River confluence. Three wetland plant communities, which provide valuable fish habitat, have been quantified by West e t a l. (1985):

* Sea g ra sses . Only a2comparatively small area of seagrass beds is present, "0.19 km , representing 0.13% of the total area in NSW. These beds are restricted to the lower most 6 km of the estuary. Summers (1997) indicates that the area of these beds decreased by 39% between 1986 and 1994.

* Mangroves. A reasonably large area of mangroves is present, 4.9 km , representing 4.6% of the total area in NSW. The mangrove communities have a relatively high richness of species. West e t a l. (1985) indicated that they extended "33 km up the estuary. My recent observations indicate that substantial patches of mangroves extend 2-4 km further upstream than this, with occasional seedlings being present 42 km upstream at Woodburn. *

* Saltmarshes. Only a comparatively small area of saltmarshes

18 2 is present, ~0.01 km , representing 0.17% of the total area in NSW. These marshes are adjacent to the estuary and restricted to the lower-most 10 km of the estuary.

West and Walford (1999) indicated that the common reed, Phragmites australis, lines much of the remaining section of the estuary. My recent observations indicate that reed occurs only in patches along the estuary, rather than a solid band along the estuary. It commences approximately 35 km upstream and is abundant by 42 km upstream at Woodburn. Its increasing abundance appears to tie in with the decreasing abundance of mangroves. Gehrke and Harris (1996) indicated that common reeds are a major source of organic debris and shelter, and they can also function as migration corridors for aquatic organisms.

Upper Richmond River arm. Very little informatiom could be located on fish habitats in this component of the estuary. Anecdotal information (Mr A. Moore pers. comm.) indicates that the estuary is in poor condition within 5 km of the tidal limit: banks are without woody vegetation and are collapsing, pools are infilling with sand and instream coarse cover (e.g. logs) is rare .

Wilsons River arm. No informatiom could be located on fish habitats in this component of the estuary.

Bungawalbin Creek arm. Anecdotal information (pers. comms. from A. Henderson and A. Moore) indicates that there is very good riparian and instream habitat structure along this arm of the estuary. Day (1994) undertook a stream health assessment of two sites located near the tidal limit of the estuary and concluded that the overall condition was good. Earley (1999) noted that the dominant wetland type in the Richmond River catchment was swamp complex which makes up 61% of the total wetland area. Most of the swamp complex is located about Bungawalbin Creek. Many lagoons are adjacent to the Bungawalbin Creek estuary.

There is strong anecdotal evidence that water quality in this arm of the estuary has declined noticeably in the last 5 years (pers. comms. from A. Henderson, L. Doust, A. Moore, G. Dodd, F. Sherwood). The key observation is that the estuary now remains turbid, instead of briefly becoming turbid during floods. Water extracted from the estuary has been observed to return overland bringing large quantities of suspended solids. I was informed that in many cases sediment-settling are not used, and in few cases naturally-occurring lagoons adjacent to the estuary are used as settling ponds.

The informants indicated that there has been a great increase in the incidence of fish with ulceration on their flanks. This observation suggests that an acidification process may be involved.

Channel geometry

Channel width. Changes in channel width along the estuary and its major arms are shown in Figure 2. Clear reductions in width occur along the middle arm of the estuary (Richmond River). The Wilsons River arm is wider than the middle arm from 70 to 105 km upstream

19 FIGURE a CHANGES IN CHANNEL WIDTH ALONG THE RICHMOND RIVER ESTUARY. Wilsons and Bungawalbin arms are included. Based on bathymetry data gathered by the NSW Public Works Department in the early 1980s. The mangrove-to- common-reed transition zone is within the 30-40 km distance interval. Key locations by distance: Broadwater = 26km, Woodburn fe 42 km, Coraki = 63 km, Casino = 113 km, Lismore = 100 km.

Channel width (m) 600 WILSONS RIVER 500

400

300

200

100

0 0 10 20 30 40 50 60 70 80 90 100 110 120

Channel width (m)

0 10 20 30 40 50 60 70 80 90 100 110 120 Distance from estuary mouth (km)

—— Mean per 5km — - S.D. per 5km Based on a water level of AHD*0 S D * standard deviation from the estuary mouth. The Bungawalbin Creek arm is considerably narrower than both other arms.

Maximum channel depth. Changes in maximum channel depth along the estuary and its major arms are shown in Figure 3. Maximum depths in the middle arm (Richmond River) start declining noticeably after 70 km upstream from the estuary mouth. Maximum depths remain high (5-6 m) in the Wilsons River arm until about 95 km upstream of the estuary mouth. In the Bungawalbin Creek arm the maximum depths vary between A to 7 m and and show no obvious decline further upstream.

Mid-channel depth. Changes in mid-channel depth along the estuary and its major arms are shown in Figure A. Middle-channel depths increase erratically along the middle arm until near the junction of the Wilsons Rivers. Both the middle arm and the Wilsons River arm have mid-channel depths decreasing erratically after this point. The rate of decrease is initially more rapid for the middle arm. In the Bungawalbin Creek arm the mid-channel depths vary between 3.5 to 6 m and show some sign of a decline further upstream.

Submersed bank slope. Changes in submersed bank slope along the estuary and its major arms are shown in Figure 5. Bank slopes increase erratically along the middle arm (Richmond River) until a point ~75 upstream of the estuary mouth. After this they decline erratically. The greatest bank slopes (5.5 m over 10 ra) were recorded at the commencement of the Wilsons River arm after which an erratic decline was apparent. Bank slopes in the Bungawalbin Creek arm vary between 3 to A.5 m over 10 m, similar to the greatest recorded in the middle arm.

Channel structure diversity index. Changes in the channel structure diversity index along the estuary and its major arms are shown in Figure 6. The index showed an erratic decrease along the middle arm (Richmond River). The maximum occurred at “18 km upstream of the estuary mouth. Another peak occurred between 50 to 60 km upstream. In the Wilsons River arm the index showed an erratic increase, peaking at “70 km upstream, then reaching a maximum at about 95 km upstream. In the Bungawalbin Creek arm the index increased noticeably after a point “70 km upstream.

The number of distinct bed anomalies. Changes in the number of distinct bed anomalies along the estuary and its major arms are shown in Figure 7. Along the middle arm (Richmond River) no anomalies were recorded up to “50 km upstream of the estuary mouth. The number then increased dramatically reaching a maximum (10 per 5 km section) near the Wilsons River confluence. The number was then sustained at 2-6 per 5 km section from 65 to “100 km upstream of the estuary mouth. In the Wilsons River arm the maximum number of anomalies occurred between 75 to 80 km upstream. The maximum number of anomalies occurred in the Bungawalbin Creek arm between 80 and 85 km upstream of the estuary mouth.

Stream-health assessment data

Riparian vegetation. The distribution of quality ratings in respect to riparian vegetation are shown in Figure 8 for four

2 0 FIGURE 3 CHANGES IN MAXIMUM CHANNEL DEPTH ALONG THE RICHMOND RIVER ESTUARY. Wilsons and Bungawalbin arms are included. Based on bathymetry data gathered by the NSW Public Works Department in the early 1980s. The mangrove-to-common-reed transition zone is within the 30-40 km distance interval. Key locations by distance: Broadwater = 26km, Woodburn = 42 km, Coraki = 63 km, Casino = 113 km, Lismore = 100 km.

Maximum depth (m)

Maximum depth (m)

Maximum depth (m)

Distance from estuary mouth (km)

—— Mean per 5km - S.D. per 5km

Based on a water level of AHD*0 S D = standard deviation FIGURE 4 CHANGES IN MID-CHANNEL DEPTH ALONG THE RICHMOND RIVER ESTUARY. Wilsons and Bungawalbin arms are included. Based on bathymetry data gathered by the NSW Public Works Department in the early 1980s. The mangrove-to- common-reed transition zone is within the 30-40 km distance interval. Key locations by distance: Broadwater = 26km, Woodburn = 42 km, Coraki = 63 km, Casino = 113 km, Lismore = 100 km.

Mid-channel depth (m)

Mid-channel depth (m)

Mid-channel depth (m)

——“ Mean per 5km - - S.D. per 5km Based on a water level of AHD-0 S D * standard deviation FIGURE S CHANGES IN SUBMERSED BANK SLOPE ALONG THE RICHMOND RIVER ESTUARY. Wilsons and Bungawalbin arms are included. Based on bathymetry data gathered by the NSW Public Works Department in the early 1980s. The mangrove-to- common-reed transition zone is within the 30-40 km distance interval. Key locations by distance: Broadwater = 26km, Woodburn = 42 km, Coraki = 63 km, Casino = 113 km, Lismore = 100 km.

Slope (depth [m] at 10 m from edge)

Slope (depth [m] at 10 m from edge)

Slope (depth [m] at 10 m from edge)

Mean per 5km ' * ■ S.D. per 5km

Based on a water level of AHD-0 Data is for both banks combined. 3 D = standard deviation FIG. 6 CHANGES IN THE CHANNEL-STRUCTURE- DIVERSITY INDEX ALONG THE RICHMOND RIVER ■ A ■ mmm CO 6 0

cd cd E CO c cd c o c o> l 5 mm m 3 .

o Q GO c O 0 0 o E 0 CO CO Z5 > c c JC TO CO n O G o CO o c CO c C CD CD CD c D £ 00 H- Q TO 1 0 < CO CD c CD CD © CO o CO O CO $ > II O 1 — CO — -C M— M— TO CO * s TO03 g o g •4—• _L_ D > CD O > O (D «- c "O C CD E E x 0 Q- O CO- 05 O x D CD CD C CO CD

i _ CC o ^ z ^ o C P* ■ MM CO Q 0) o E L. LL o cO O c ■ M i CC o £ HI n r (0 CD £ to 2 H i CD Z> X c Z H 3 £D LU 0 X z otf - Bungawalbin Wilsons i- o V) . Z < O _ _ w co co = HI HI O J . Z < > < 2 X

X o < estuary (km) from mouth Distance “ Richmond — o © < ^ < & 5 > 0 Q 111 © r r HI © “ “ CD CO £ © c o T3 © © © CD FIGURE 2 THE DISTRIBUTION OF QUALITY RATINGS IN FOUR SECTIONS OF THE RICHMOND RIVER ESTUARY: RIPARIAN VEGETATION. Data from the Stream Health Assessment of Bird (1997).

Upper Richmond

Very poor Poor Moderate Good Very good

WUeone River

Percentage (%} 100

80

00

40

20 0 Very poor Poor Moderate Good Very good

Bung&wcJbln Greek Percentage (%} 100

80

00

40

20

0 Very poor Poor Moderate Good Very good

Lower Richmond

Percentage (%) 100 80 eo

40

20 0l = e ^ Very poor Poor Moderate Good Very good Quality ratine sections of the estuary. The estuary component which tended to have the highest quality ratings was the Bungawalbin 'Creek arm (45% very good). The lower Richmond (42% very poor) followed by the upper Richmond (65% poor) tended to have the lowest quality ratings.

Aquatic habitat. The distribution of quality ratings in respect to aquatic habitat are shown in Figure 9 for four sections of the estuary. The estuary component which tended to have the highest quality ratings was the Bungawalbin Creek arm (85% good). The upper Richmond (62% poor) and the lower Richmond (26% very poor) tended to have the lowest quality ratings.

Aquatic vegetation. The distribution of quality ratings in respect to aquatic vegetation are shown in Figure 10 for four sections of the estuary. Very poor ratings dominated all the components of the estuary. The Bungawalbin Creek arm had a slightly better rating distribution compared to the other components of the estuary.

Overall condition. The distribution of quality ratings in respect to the overall condition are shown in Figure 11 for four sections of the estuary. The estuary component which tended to have the highest quality ratings was the Bungawalbin Creek arm (80% good). The upper Richmond (63% poor) followed by the lower Richmond (26% very poor) tended to have the lowest quality ratings.

Riparian vegetation (aerial photographic data)

The length of riparian canopy along the major arms of the estuary are shown in Figure 12 for four canopy-width categories.

Trace category. Along the middle arm (Richmond River) there appears to be four broad peaks: from 3 to 44 km upstream with a maximum at 23 km (corresponds to fringing mangroves - see Figure 13), between 45 and 63 km with a maximum at 62 km which is just before the Wilsons River confluence, between 64 and 99 km with a maximum at 71 km, and between 106 and 113 km with a maximum at 110 km. The Wilsons River arm, in comparison to the middle arm, is more consistently lined with this category of riparian vegetation. The Bungawalbin River arm, particularly after 64 km upstream, is consistently lined with riparian vegetation. The amount of the bank lined is greater than what occurs along either the middle or the Wilsons River arms.

25 m category. Along the middle arm (Richmond River) the first broad peak associated with fringing mangroves becomes more conspicuous under this category. Its upper limit drops back to “36 km upstream of the estuary mouth. Compared with the previous category, the more upstream broad peaks along the middle arm recede considerably, as does the coverage along the Wilsons River arm. The Bungawalbin Creek arm, beyond 65 km upstream, commences to stand out because of the consistent large amount of bank lined.

50 m category. Along the middle arm (Richmond River) the first broad peak associated with fringing mangroves becomes more conspicuous. The more upstream broad peaks along the middle arm

21 FIGURE 9 THE DISTRIBUTION OF QUALITY RATINGS IN FOUR SECTIONS OF THE RICHMOND RIVER ESTUARY: AQUATIC HABITAT. Data from the Stream Health Assessment of Bird (1997).

Upper Richmond

Very poor Poor Moderate Good Very good

Wlbona River

Percentage (%} 100

80

00

40

20

0 Very poor Poor Moderate Good Very good

Buogawedbki Creek

100

80

00

40

20

O Very poor Poor Moderate Good Very good

Lower Richmond Percentage (%} 100

80 80

40

20

O Very poor Poor Moderate Good Very good Quality rating FIGURE 10 THE DISTRIBUTION OF QUALITY RATINGS IN FOUR SECTIONS OF THE RICHMOND RIVER ESTUARY: AQUATIC VEGETATION. Data from the Stream Health Assessment of Bird (1997).

Upper Richmond

Percentage (%)

Very poor Poor Moderate Good Very Qood

W ilsons River

Percentage (%)

Very poor Poor Moderate Good Very good

Bungawedbin Creek

Percentage (%)

Lower Richmond

Percentage (%)

Quality rating FIGURE II THE DISTRIBUTION OF QUALITY RATINGS IN FOUR SECTIONS OF THE RICHMOND RIVER ESTUARY: OVERALL CONDTION. Data from the Stream Health Assessment of Bird (1997).

Upper Richmond

Very poor Poor Moderate Good Very good

Wtteons River

Very poor Poor Moderate Good Very good

Bungawefcln Crook

Very poor Poor Moderate Good Very good

Lower Richmond

Very poor Poor Moderate Good Very good Quality rating FIG. 12 LENGTH OF RIPARIAN CANOPY ALONG THE RICHMOND RIVER ESTUARY IN ONE KM INTERVALS: four width categories.

Trace category

Pipanan length (km) per river Kilometre

26 m category

Riparian lenQth (km) par rlvar kllomatra

60 m category

100 m category

— Richmond B vngevelbn Wilaona have virtually disappeared, as has the coverage along the Wilsons River arm. The Bungawalbin Creek arm, beyond 73 km upstream, stands out because of the consistent large amount of bank 1ined.

100 m category. Along the middle arm (Richmond River) the first broad peak associated with fringing mangroves weakens noticeably under this category. The more upstream broad peaks along the middle arm have virtually disappeared, as has the coverage along the Wilsons River arm. The Bungawalbin Creek arm, particularly between 81 and 89 km upstream, stands out because of the consistent large amount of bank lined.

Mangrove canopy-width. The length of mangrove canopy along the major arms of the estuary are shown in Figure 13 for four canopy- width categories. The coverage of all categories reduced noticeably between 30 and A2 km upstream of the estuary mouth. This decline may be related to an increase in submersed — bank^ slopes along this section of the estuary (see Figure 5).

3.1.2 Fish fauna

Available information on the fish fauna and fisheries of the major components of the Richmond River estuary are summarised in this section. As before, the border between the lower and the upper Richmond River components of the middle arm of the estuary, is taken to be at the Wilsons River arm confluence, 63 km upstream from the estuary mouth. The majority of commercial, recreational, and aquaculture-based fishing activities are concentrated in the lower Richmond River component. Coraki at 62 km upstream is generally considered to be the upstream limit of trawling in the estuary (R. Williams pers. comm., NSW Fisheries). Information on these fisheries is given primary attention in the lower Richmond River component.

Lower Richmond River (lower middle arm)

Commercial fishery. Based on work undertaken in 1988-89, West (1993a & b) indicated that the fisheries resources of the Richmond River estuary were in comparatively good overall condition, taking into account the history of harvesting and environmental changes which have occurred. In recent years (1982/83-91/92) the total production from the estuary averaged ~210 tonnes, representing only 3.2% of the total production from NSW estuaries (data from Pease and Grinberg [1995]).

In 1991-92 the production was 239 tonnes and was broken up as follows:

* finfish = 161 t (sea mullet 81, 37, luderick 13, flat-tail mullet 11, sand whiting 8, bream [black and yellowfin] 3, mulloway 3) * crustaceans = 78 t (school prawns 72, mud A)

Pease and Scribner (1993) found that the total value of the catch in 1990/91 was $ 733,283, representing 3.9% of the value of the total estuarine catch in NSW. In 1996-97 the total value of the catch was $8A9,A67 (NSW Fisheries unpublished data).

22 FIO.B LENGTH OF MANGROVE CANOPY ALONG THE RICHMOND RIVER ESTUARY IN ONE KILOMETRE INTERVALS: Four categories CO CM O CM CO O LD O ID CM CO O LD O ID O LD ID 'sh O ID ID CO O

Distance to river mouth (river km)

Trace ~~- 25 m — 50 m —— 100 m Recreational fishery. Both Nihill (1994) and West and Gordon (1994) have investigated recreational fisheries in the lower Richmond River estuary. All the information below is sourced from the more detailed study of West and Gordon (1994). The estimated recreational-angling effort from March 1988 to February 1989 in the lower-most 20 km of the estuary was 223304 hours. When adjustment is made for the area of the estuary, the effort is 238 h/ha/yr, the highest estimate for any NSW estuary.

The estimated breakup of main elements of the catch during the study of West and Gordon (1994) is given below:

* yellowfin bream = 12-19 t/year (commercial = 1 t/year) * dusky flathead = 5-10 t/year (comm = 2 t/year) * sand whiting = 1-4 t/year (comm = 3 t/year) * luderick = 1-4 t/year (comm = 12 t/year) * mulloway = <1 t/year (comm = 2 t/year) * tailor = 1-5 t/year (comm = <1 t/year)

Bully mullet were seldom utilised by the recreational anglers yet they represented 77% of the total finfish commercial catch (148,972 tonnes) in the estuary in 1988/89 (data from Pease and Grinberg [1995]).

Aquaculture. The primary aquaculture venture in the Richmond River estuary is the production of Sydney rock oysters. The areas of production are from 2 to 10 km by estuary upstream from the estuary mouth (NSW Fisheries, unpublished data).

The production of oysters from the estuary is quite low. In recent years (1982/83-91/92 inclusive) the production averaged “16 t/year and was quite variable. This represented 0.2 % of the total NSW production (data presented by Pease and Grinberg [1995]).

In 1990/91 the value of the oyster catch from the estuary was $103410, representing 0.4% of the value of the total production from NSW (data presented in Pease and Scribner [1993]).

Fish studies in the mainchannel. There have been at least two student projects concerning the lower mainchannel environment. Cody (1995) examined the fish assemblages associated with seagrass habitats, and Peverell (1996) developed a finfish database for the lower Richmond River estuary near Ballina.

Bishop (1999c) reported anecdotal information indicating that European carp are occasionally captured by commercial fishermen in the lower estuary. It was further indicated that golden perch, a native freshwater fish species translocated from the Murray- system, has been captured by fishers in the estuary near Ballina.

By far the most detailed information on the fish fauna of the lower estuary arises from work undertaken by NSW Fisheries in 1988-90 (originally reported in West [1993a&b], and later in West and Walford [1999]). As detailed in Section 1.5, the work involved taking replicated, quarterly samples by trawling over a two-year period along the subtidal environment in the main deepwater channel of the estuary. In all 17 systematically-

23 selected longitudinal sites were examined, 14 of which were in the lower Richmond River component of the estuary.

West and Walford (1999) indicated that 107,174 fish across 66 species were captured from the estuary. The number species collected decreased with increasing distance from the mouth of the estuary: 36 species near Ballina, and down to 16 species near Coraki. Based on the analysis of the species composition of samples, three zones were recognised along the estuary:

* up to "15 km upstream of the estuary mouth (3 sites); referred to as the marine section:

- the most distinguishing species were: + long tailed catfish + banded toad + sand whiting

* from "15 to "50 km upstream (8 sites); referred to as the brackish section:

- the most distinguishing species were: + forktailed catfish + narrow banded sole + mulloway

* beyond "50 km upstream (6 sites, 3 of which were in the lower Richmond component of the estuary); referred to as the freshwater section:

- the most-distinguishing (+), exclusively-caught (£), or predominately-caught (@) species were:

+ Australian bass + freshwater herring + bullrout £ eeltailed catfish ( Tandanus sp.A) £ freshwater mullet @ bully mullet (primarily juveniles) @ flatheaded gudgeon

Fish studies in estuarine-reaches of tributaries that are connected to the lower estuary. There has been at least one student study on the fish fauna within a tributary of the lower estuary. Graham (1989) examined fish utilisation in the Newrybar flood mitigation drain. This drain enters North Creek which in turn enters the lower estuary ~2km upstream from the estuary mouth. A range of netting techniques were used to sample four sites along the drain and one site in North Creek. These sites were sampled once in September/October 1989. Four species of primarily-freshwater fish species were recorded in the upper reaches of this tributary.

Bishop (1999c) provided a summary (Table 2) of recorded occurrences of fish in four creek-long sections of Emigrant Creek, including the upper estuary. The summary arose from a consolidation of his work, surveying by NSW Fisheries (partially reported by Harris and Gehrke [1997]), and anecdotal records. It is clear that a considerable range of native primarily-freshwater

24 TABLE 2 THE RECORDED OCCURRENCE OF FISH TAXA IN FOUR CREEK-LONG SECTIONS OF EMIGRANT CREEK. From Bishop (1999c). The alpha coding allows cross referencing to source references.

Fish taxa Section of Emigrant Creek

Emigrant Emigrant Killen Estuary: Ck. Dam Creek Falls tidal feeder Dam down to limit streams tidal down to limit 7km

Native primarily-freshwater

Longfinned P N Freshwater herring - - P P, N Eeltailed catfish Du, Me Do,F,P P N Duboulay's rainbowfish - - P P ,N Pacific blue-eye - - - P , N Agassizi's perchlet - - - P, N Bullrout - - - N Eastern freshwater cod - A A A Australian bass - F,P P P , N Bully mullet - - P P? , N Freshwater mullet - - - P? , N Cox's gudgeon - - P P Striped gudgeon - - P P , N Empire gudgeon - - P P , N Firetailed gudgeon - F, P P P Flathead gudgeon F, P P , N

Native primarily-estuarine fishes:

Snub-nosed garfish N Garfish sp? P Estuary perch N Golden trevally N Yellowfin bream N Flat-tail mullet N

Translocated native fishes: Golden perch Do,F,P A A Silver perch Do P

Alien/pest fishes:

Gambus ia P P P, N European carp F,P P P Redfin perch A

Reference coding: Do = Doohan (1989), Du = Dunk (1991), F = Faragher e t a l . (1993), M = Melville (1991), N = NSW Fisheries as partially derived from Harris and Gehrke (1997) , P = Bishop's work; A = anecdotol evidence derived from Bishop's work. - = not recorded fish species have been recorded in the upper estuary of Emigrant Creek. A notable occurrence is the Eastern freshwater cod which has a high conservation value.

Invertebrate studies. There have been at least three studies on benthic macroinvertebrates in sand and/or seagrass habitats in the lower estuary (Dwyer [1996], Currie (1998) and Reid (1992).

Upper Richmond River arm

The most extensive fish surveying work undertaken in this arm is the trawling work of NSW Fisheries, originally reported in West [1993a&b], and later in West and Walford (1999). Three sites were sampled in this arm and they ranged from 65 to 71 km by estuary upstream from the estuary mouth, i.e. the most- downstream portion of the arm. Only 17 species were captured. Based on the species composition of the samples, the area was considered to be typical of a freshwater zone. The most- distinguishing species of this zone were Australian bass, freshwater herring and bullrout. Species exclusively-caught in reasonable numbers in the zone were eeltailed catfish ( Tandanus sp.A) and freshwater mullet. Species predominately-caught in the zone were bully mullet (primarily juveniles) and flatheaded gudgeons.

Data from the Australian Museum database indicates that the Oxleyan Pygmy perch, Nannoperca oxleyana, has been recorded in this arm of the estuary at a point ~97 km upstream of the estuary mouth (10 km upstream of Tatham). The record is dated July 1929. The record appears to be of high quality as: eight specimens were kept as paratypes, a distinguished ichthyologist (Mr G. Whitley) made the identifications, and other species from the same collection had the same spatial reference (latitude and longitude). The Oxleyan pygmy perch has a high conservation value. It may have been captured in beds of ribbon weed which are still apparent today (P. Gooley pers. comm.).

Anecdotal evidence (pers. comms. from Dr S. Rowland [NSW Fisheries], and Mr G. Dodd) indicates that another high- conservation value species, the Eastern freshwater cod, occurs close to the tidal limit near Casino. Mr G. Dodd indirectly reported the capture of a large Cod ("A kg) from immediately below Grays Falls in August 1997. Grays Falls is the tidal barrier for the estuary arm.

Other anecdotal records indicate the following occurrences:

* bull sharks (Carcharhinus leucas) are abundant near Grays Falls and further downstream (A. Moore and G. Dodd pers. comm.); they are targeted by anglers in summer * tarpon (Megalops cyprinoides) occur downstream of Grays Falls (G. Dodd pers. comm.) * European carp occur about the tidal limit (G. Dodd pers. comm.) * Dusky flathead occur about the tidal limit (A. Moore pers. comm.) * Yellowfin Bream occur about the tidal limit (A. Moore pers. comm.)

25 * a very good Australian bass population (mainly juveniles) occurs about the tidal limit (G. Dodd pers. comm.)

Wilsons River arm

The most extensive fish surveying work undertaken in this arm is that by NSW Fisheries as a part of the NSW Rivers Survey (reported by Harris and Gehrke [1997]). Four comprehensive samples were taken between August 1994 and January 1996 at a site in Leycester Creek ~2 km upstream from the Wilsons River confluence near Lismore. The site is well below the tidal limit (MHL 1998) and is ~103 km by estuary upstream of the estuary mouth. A list of species collected at this site is given in Table 3. In all 17 species were recorded, 16 of which were native primarily-freshwater species.

Austin (1992) has additionally recorded freshwater herring in the Wilsons River downstream of Lismore. Anecdotal records (A. Moore pers. comm.) indicate that bull sharks commonly occur in the Wilsons River near Lismore, and that dusky flathead are present in the tidal-barrier at the head of the Wilsons River arm of the estuary (immediately downstream of the Rd. crossing).

Bungawalbin Creek arm

There have been at least three student studies in or near the upper reaches of this arm of the estuary. In 1994 Day (1994) sampled 2 sites a short distance downstream of the tidal limit. Pearce (1994) sampled 4 sites up to 6 km downstream of the tidal limit from 29th September to 20th October 1994. Stewart (1998) sampled 3 lagoons adjacent to the estuary with limited methods on 1st to 11th September 1998. Seventeen native primarily-freshwater species were collected in the estuary along with one alien/pest species (Table 4). A notable absence was the freshwater herring. Six native primarily-freshwater species and one alien/pest species were collected in the adjacent lagoons.

Anecdotal information from a considerable number of sources indicates a major decline in the Australian bass fishery in the Bungawalbin arm of the estuary in recent years:

* 3 years ago a crash commenced; instead of 14-17 bass in 4 hrs fishing, it has reduced to 1-2 bass in 4 hrs (G. Dodd pers. comm.) * 4 years ago a crash commenced; the last good recruitment was in 95/96; there use to be a very good bass fishery (A. Moore pers. comm.) * 3-5 years ago the bass fishery commenced a decline (F. Sherwood pers. comm.) * 5 years ago a major crash in the bass population started, now they are nearly extinct; it use to be one of the best bass fisheries in NSW (L. Doust pers. comm.) * 10 years ago a dramatic decline in the bass population commenced (A. Henderson pers. comm.)

A. Henderson (pers. comm.) indicated that mullet, catfish, shortfinned eels and school prawns have also shown a major

26 TABLE 3 LIST OF FISH SPECIES RECORDED BY HARRIS AND GEHRKE (1997) IN THE ESTUARINE SECTION OF LEYCESTER CREEK, A BRANCH OF THE WILSONS RIVER ESTUARY ARM.

Fish taxa

Native primarily-freshvater fishes:

Longfinned eel

Freshwater herring

Eeltailed catfish ( Tandanus sp.A)

Duboulay's rainbowfish

Pacific blue-eye

Olive perchlet

Bullrout

Australian bass

Bully mullet Freshwater mullet

Striped gudgeon Empire gudgeon Firetailed gudgeon Flathead gudgeon Undescribed flathead gudgeon

Native primarilv-estuarine fishes:

Fork tailed catfish TABLE A LIST OF FISH SPECIES RECORDED IN THE BUNGAWALBIN CREEK ARM OF THE RICHMOND RIVER ESTUARY BY PEARCE (1994) AND DAY (199A). Species recorded by Stewart (1998) in lagoons adjacent to the estuary are also indicated.

Fish taxa Estuary Lagoons (Stewart Pearce (199A) Day (199A ) 1998 )

Native primarily-freshwater fishes:

Longfinned eel + +

Australian smelt + + +

Eeltailed catfish + +

Crimsonspotted rainbowfish + Duboulay's rainbowfish +

Pacific blue-eye + +

Agassiz's perchlet + +

Bullrout + +

Australian bass + +

Bully mullet + Freshwater mullet + +

Striped gudgeon + + + Cox's gudgeon + + + Empire gudgeon + + + Firetailed gudgeon + + + Western cap gudgeon + + + Flathead gudgeon + +

Alien/pest fishes:

Gambusia + + +

+ recorded decline over the last ten years. As noted earlier, a number of informants (G. Dodd, F. Sherwood) indicated that there has been a great increase in the incidence of fish with ulceration on their flanks.

There are indirect reports that Eastern freshwater cod occurred in the system in the past (L. Doust quoting B. East). Recently there have been indirect, unsubstantiated reports of juvenile cod ~10 cm in length being present (A. Moore pers. comm.). However the informant suggests possible identification difficulties, noting that large gudgeons resemble juvenile cod. I have also located a student thesis which indirectly reports Cod from the Bungawalbin estuary (the thesis is yet to be released from intellectual property restrictions).

3.2 The identification of high-value and vulnerable components

The consideration of high value habitat and fish-fauna components is limited to those occuring further upstream than 50 km upstream of the estuary mouth. West and Walford (1999) indicated that the freshwater zone commences in this area. Saline water may move upstream into this area as a result of freshwater extraction further up the estuary.

3.2.1 Fish habitat

Based on channel geometry parameters (channel-structure diversity index = CSDI; number of distinct bed anomalies = NDBA), the following areas are identified as potentially having high value:

* 50-60 km upstream on the middle arm (peak in CSDI) * 65-70 km upstream on the Wilsons River arm (peak in CSDI) * 95-105 km upstream on the Wilsons River arm (peak in CSDI) * >80 km upstream on the Bungawalbin Creek arm (peak in CSDI)

* 60-65 km upstream on the middle arm (peak in NDBA) * 85-90 km upstream on the middle arm (peak in NDBA) * 75-80 km upstream on the Wilsons River (peak in NDBA)

Based on the stream health data of Bird (1997), the following components of the estuary are identified as potentially having high value:

* Bungawalbin Creek arm (riparian vegetation ratings) * Bungawalbin Creek arm (aquatic habitat ratings) * Bungawalbin Creek arm (overall condition ratings)

Based on the riparian canopy-width data derived from aerial photographs, the following areas of the estuary are identified as potentially having high value:

* Bungawalbin Creek arm > 65 km upstream (25 m category) * Bungawalbin Creek arm > 73 km upstream (50 m category) * Bungawalbin Creek arm 81-89 km upstream (100 m category)

To a lesser extent these areas potentially have high value in respect to riparian canopy-width data:

27 * middle arm 60-62 km upstream (25 m category) * middle arm 74-83 km upstream (25 m category) * middle arm “110 km upstream (25 m category) * Wilsons River arm 73-74 km upstream (25 m category) * Wilsons River arm ~80 km upstream (25 m category) * Wilsons River arm 92-104 km upstream (25 m category) * Wilsons River arm 112-113 km upstream (25 m category)

The Bungawalbin Creek arm stands out as a high value component of the Richmond River estuary.

3.2.2 Fish fauna

The species occurring in the freshwater zone of the estuary with high-conservation value are:

* Oxleyan pygmy perch (Nannoperca oxleyana) ; listed as an under the provisions of the Fisheries Management Amendment Act 1997 - high-quality historical record in the upper Richmond River arm of the estuary

* Eastern freshwater cod (Maccullochella ikea); listed as an endangered species under the provisions of the Fisheries Management Amendment Act 1997 - low-quality (indirect) records in the upper Richmond River and Bungawalbin Creek arms of the estuary including Emigrant Creek and Newrybar flood mitigation drain

* Eeltailed catfish (Tandanus unident. sp.A) of potentially high conservation value; genetic evidence (Jerry and Woodland 1997) indicates the catfish found in the Richmond River system are possibly a new species or subspecies - high quality records in all arms of the estuary including Emigrant Creek and Newrybar flood mitigation drain

The species occurring predominately in the freshwater zone of the estuary with only high-recreational value is:

* Australian bass (Macquaria novemaculeata); listed by West and Walford (1999) - high quality records in all arms of the estuary including Emigrant Creek

West and Walford (1999) listed 8 species which have both commercial and recreational value. All of these were predominately caught in either the marine (7 species) or brackish zone (1 species). Only 3 of these were occasionally caught in the freshwater zone: dusky flathead (4% of catch), tailor (3%), and yellowfin bream (2%).

West and Walford (1999) listed 4 species which only have commercial value. Of these only the roach, bully mullet and freshwater mullet were recorded in the freshwater zone. The latter two species were predominately captured in the freshwater zone. Accordingly, species which only have a high commercial value and were predominately captured in the freshwater zone are: *

* Bully mullet (Mugil cephalus); listed by West and Walford

28 (1999) high quality records in all arms of the estuary including Emigrant Creek (mainly juveniles)

* Freshwater mullet (Myxus p e ta r d i); listed by West and Walford (1999) high quality records in all arms of the estuary including Emigrant Creek (mainly juveniles)

Another species which has a high commercial value (through aquaculture), yet it only occurs in the lower Richmond River arm of the estuary, is the Sydney rock oyster (Saccostrea commercialls) . This species has been included for consideration as it may be vulnerable to increases in salinity through a biofouling mechanism involving marine fauna. j ■•..3—Identifying salinity-exposures likely to cause change j. •. 3 • 1_Li terature/discussion derived indications of salinity tolerances

Oxleyan__pygmy perch. McDowall ( 1996) indicates that this species is found in waters with very low conductivity, <200 uS/cm, or using the conversion factor in Hart e t a l. (1991), 0.14 ppt total soluable salts. Arthington (1996) presented an array 16 conductivity values of waters in which the Perch had been collected. One quite high value was recorded, 2148 uS/cm (or 1.5 ppt TDS), but this was potentially an anomally as all other values were below 352 uS/cm (0.24 ppt TDS). The mean conductivity value without the potential anomally was 134 uS/cm (0.07 ppt TDS). An indicator of salinity tolerance (1ST) based on this mean, and the variability about the mean, was devised as follows: 1ST = mean + standard deviation of the mean

The determined value was 171 uS/cm, or 0.12 ppt TDS. It is stressed that this is only an indicator of salinity tolerance because absence in the field may be due to factors besides salinity tolerance. Additionally, the associated salinities may simply reflect the distribution of salinities sampled. Even if salinity limits were consistent, this may be reflecting indirect effects of salinity, such as habitat loss or increased competition or predation from salt-tolerant species. Arthington (1996) noted that the Perch may be susceptible to increased competition and predation pressure.

Eastern freshwater cod. Dr S. Rowland (pers. comm.) has kept Cod in 5 ppt saline water for 2-3 months with no apparent harmful effects (the effects were potentially beneficial as stress can be reduced by removing the cilliate parasite Chilodonella hexa stica which affects the skin and gills). Historically it appears (Rowland pers. comm.) that Cod were not recorded in the Clarence River downstream of Grafton where salinity does not generally exceed 1 ppt (Howland 1998). Considering the above two points, it appears that physiological stress caused by salinity intolerance is unlikely to be an important factor for the Cod. Other associated factors which could cause the apparent downstream distribution limit could be:

29 * increased competition with estuarine fish * increased predation from estuarine fish * the Cod is essentially a riverine fish and accordingly seeks such habitat

The best-available indicator of the presence of Cod is taken to be 1 ppt salinity. The impacts of salinity on Cod are likely to be indirect.

Eeltailed catfish. Nothing regarding salinity tolerances was located on fish of the genus Tandanus.

Australian bass: spawning, egg hatching & initial survival of larvae. Harris (1986) indicated that bass in areas of the estuary with salinities ranging from 8 to 14 ppt. This was based on the location of recently spent fish, the distribution of larvae and sperm viability trials. The upper limit (14 ppt) is questionable as bass have only been reliably recorded in salinities up to 13 ppt (Harris 1986).

Through laboratory testing, Van der Wal (1985) found that optimal salinities for the hatching of eggs and survival of larvae was 25 to 35 ppt. Also through laboratory work, Battaglene and Talbot (1993) found optimal larval survival to have a wider salinity range, 10 - 35 ppt. Optimal salinity for swim bladder inflation of the larvae was found by these latter authors to be 15 - 35 ppt.

Australian bass non-spawning period: older larvae, juveniles and adult males. Harris (1987) indicated that most male bass remain in the estuary while females predominate in upland lotic habitats. While adult bass have been reliably recorded in salinities only up to 13 ppt (Harris 1986), their common limit outside spawning period is 5 ppt (Harris pers. comm.). There is little doubt that this apparent salinity range preference results in a number of advantages for the bass, the most important being reduced interspecies competition and predation by species typically found in the middle to lower estuary (e.g. river whaler sharks, dewfish, bream, flathead, blackfish, etc). A similar advantage would result for larval and juvenile bass occupying the upper estuary. Such advantages gained by inhabiting the upper estuary have frequently been reported for estuaries outside of Australia (e.g. Drinkwater and Frank 1994).

Larval and juvenile bass are dependent on beds of macrophytes, mostly ribbon grass, Vallisneria spp., and common reeds, Phragmites australis, to which they recruit while migrating upstream in the estuary (Harris 1983, 1985, 1986 and 1988). The beds provide shelter from predators, an abundant food supply, and may be affected by a range of factors. Salinity changes could be important factors influencing these beds.

Harris (pers. comm.; cited in Bishop [1995]) indicated that ribbon grass can tolerate 5 ppt salinity for one to two months. Sainty and Jacobs (1981) indicated that this species thrives at 1.5 ppt.

In respect to the common reed, Harris (pers. comm.; cited in Bishop [1995]) indicated that it could tolerate 25 ppt salinity

30 for one to two months. Sainty and Jacobs (1981) indicated that sea-strength salinity (35 ppt) for a few days will severely damage this species. Dalby-Ball, Sainty & Jacobs (1999; Appendix A to the Main Report) indicated that there is a very wide range of reported salinity tolerances for this species: 5-65 ppt, but mostly 5-25 ppt. Within the Richmond estuary the common reed greatly reduces in abundance in a transition zone 30 to 40 km upstream of the estuary's mouth. In the midpoint of this zone the salinity regime under existing conditions (data supplied by WRL, see Main Report) is characterised as:

* below 5 ppt 90% of the time * below 1.5 ppt 70% of the time * at ~0 ppt 50% of the time

It is clear that common reeds in the Richmond River estuary are rarely exposed to salinity levels of 25 ppt or greater. A reasonably-conservative indicative 'safe' salinity level for this species in the system appears to be in the order of 5 ppt.

Bully mullet. McDowall (1996) noted that this species spawns at sea and that the freshwater phase is not obligatory. Merrick and Schmida (1984) indicated that lowered salinities definitely seemed to attract juveniles. Correspondingly, Pollard and Hannan (1994) classified the bully mullet as euryhaline. Based on direct salinity tolerances, it is anticipated that bully mullet would be minimally affected by even moderate changes in the salinity structure of the estuary.

Freshwater mullet. McDowall (1996) noted that mature adults move down to the estuaries and sea to spawn. Merrick and Schmida (1984) noted that spawning occurs in lower salinities. As for bully mullet when considering direct salinity tolerances, it is anticipated that freshwater mullet would be minimally affected by even moderate changes in the salinity structure of the estuary.

General maintenance of freshwater ecosystems. Hart e t a l. (1991) and ANZECC (1992) indicated that direct adverse biological effects are likely to occur in freshwater ecosystems if salinity is increased to around 1 ppt. A summary of information relating to salinity tolerances of non-high-value freshwater fish found in the Richmond River estuary is given in Table 5.

Sydney rock oyster. Nell and Holliday (1988) showed through laboratory experimentation that oyster larvae had highest growth rates at salinities of 23 to 39 ppt, and highest survival rates at 27 - 39 ppt. Two groups of oyster spat with initial weights of 1.3 mg and 0.61 g grew best at salinities of 25 to 35 ppt and 20 - 40 ppt, respectively. Salinity had no significant effect on the survival of the spat.

Higher salinities are associated with the incidence of 'winter mortality' disease which affects this species of oyster. The disease was first described by Roughley (1926) and the causative , Mikrocytos roughleyiy was first described by Farley et a l. (1988). The disease is prevalent at 30 to 35 ppt salinity and is unknown at lower salinities (Farley et a l. 1988). It is unknown whether the oysters become susceptible to the disease through stress caused by higher salinities, or the disease

31 TABLE 5 SUMMARY OF INFORMATION RELATING TO SALINITY TOLERANCES OF NON-HIGH-VALUE FRESHWATER FISH FOUND IN THE RICHMOND ESTUARY.

Native fish species:

* Freshwater herring: can live in freshwater or seawater (data summarised in Koehn & O'Conner [1990])

* Fork-tailed catfish: can complete its life cycle in freshwater; populations in estuarine and marine water are reported to undertake extensive anadromous migrations (McDowall 1996)

* Southern blue-eye: observed in the field to 10 ppt (data summarised in Hart e t a l. [1991])

* Duboulay's rainbowfish: laboratory 4 day LC f°r eggs was 22 ppt and fry was 21 ppt (data summarised m Hart e t al. [1991])

* Bullrout: thought to be catadromous (McDowall 1996)

* Striped gudgeons: juveniles found in estuaries (data summarised in Koehn & O'Conner [1990])

* Flatheaded gudgeons: > 20 ppt, been found up to 7.3 ppt, frequently occurs in estuaries, euryhaline (data summarised in Koehn & O'Conner [1990])

* Firetail gudgeon: restricted to freshwaters (McDowall 1996)

* Empire gudgeon: schools of juveniles are often found in estuaries (McDowall 1996) *

* Carp gudgeon (western): been found in 8.8 ppt, 4 day LC5Q 38 ppt (data summarised in Koehn & O'Conner [1990])

Alien/pest fish species:

* Gambusia: tolerates a wide range of salinities, from pure freshwater to full marine conditions (McDowall [1996], Merrick and Schmida [1984]) organism is more active under these conditions. The feature that optimal growth rates for oysters occur at higher salinities tends to eliminate the possibility of the above stress-related mechanism. Salinity may not be a conditioning environmental factor at all, but instead, be correlated to such a factor. As the disease occurs in winter, and is only present in the southern range of the oyster, low water temperatures could be implicated as a conditioning environmental factor. Wolf (1967) indicated that the disease has occurred from Victor Harbour in S.A. to the estuary on the NSW mid-north coast. As oysters are commonly transported between estuaries, it is possibile that the causative organism could reach the Richmond River estuary.

Higher salinities are frequently associated with the 'fouling' of oyster racks by marine animals such as G aleolaria , bryozoans, barnacles and cunjevoi (Ciona intestinalis ). Fouling is most prevalent in winter months and is more of a problem with floating raft cultures (I. Smith pers. comm., NSW Fisheries). The salinities at which fouling problems commence have not been accurately determined. Bishop (1995) indicated that it could be reasonably assumed that an indicative level for the commencement would be about 30 ppt. For the present study I 'revisted' this value through discussions with Dr R. De Nys of the University of N.S.W. Biofouling Research Group. It was indicated that 20 ppt salinity would be a much better indicator of a threshold of increased biofouling problems.

Summary of indicative salinity limits. The following six salinity values were considered as indicative of some important ecological limits in the estuary as relevant to fish:

0.12 ppt: upper limit for Oxleyan pygmy perch

1 ppt: upper limit for Eastern freshwater cod (indirect impacts) and the maintenance of freshwater ecosystems

5 ppt: upper limit for adult Australian bass outside the spawning season as well as for ribbon grass which is important the shelter of larval and juvenile bass; possibly a suitable limit for common reeds which also provide important shelter.

8-13 ppt: lower and upper limit for adult Australian bass during the spawning season.

20 ppt: indicative limit for the Sydney rock oyster above which there is an increased chance of fouling of attachment substrates by marine animals.

3.3.2 Analysis of data gathered along salinity gradients

The following estuary-long data sets were investigated, and where possible, examined in relation to parallel sets of salinity data. Optimal examination involved accessing data at the individual sample level, as opposed to multi-sample data summaried per site, or across sites (suboptimal examination).

32 Dataset A1:_fish communities from the Emigrant Creek estuary

This dataset arose from the work of Bishop (1999c) undertaken in March/April 1999 within the upper Emigrant Creek estuary. In an attempt to locate Eastern freshwater cod and Oxleyan pygmy perch, intensive sampling using lure fishing and bait trapping was undertaken along the most-upstream 7 km of the estuary.

The lure fishing was restricted to daylight hours and 30 minutes of fishing at a given point was the standard fishing (effort) unit used. Higly-skilled lure fishermen were hired for the work. A minimum of 12 units were randomly deployed within each 1 km section of the estuary. A total of five fish species were recorded, not including the Eastern freshwater cod. Australian bass was the most commonly caught species and it was most abundant within 3 km of the tidal limit.

Bait trapping, which targeted the Oxleyan pygmy perch, was restricted to daylight hours and one hour of trapping at a given point was the standard trapping (effort) unit used. A minimum of 30, but generally 50 units were randomly deployed within each 1 km section of the estuary. A total of nine species were captured in the estuary not including the Oxleyan pygmy perch. Catches were dominated by the native empire gudgeon followed by the alien pest fish Gambusia. The empire gudgeon was most abundant in the upper 3 km of the estuary while Gambusia had two obvious peaks in abundance, at 1-2 km and 6-7 km downstream of the tidal limit. The native southern blue-eye was the next-most abundant species and its distribution along the estuary was similar to that of the Gambusia. Three fish species appeared to be restricted to the upper two kilometres of the estuary: Duboulay's rainbowfish, the Agassizi's perchlet and Cox's gudgeon. These species are typically associated with flowing streams upstream of estuaries.

The patterns in species abundance and richness were not related to salinity as all waters were fresh at the time of sampling. An initial attempt was made to link patterns in with available habitat data. The number of species collected in of the seven estuary sections is given

Bait Lure trapping fishing

Section ECE1 6 2 Section ECE2 10 A Section ECE3 A 3 Section ECEA 6 1 Section EGE5 1 2 Section ECE6 3 2 Section ECE7 6 3

These variables were correlated (Pearson's product-moment) with the corresponding mean values of the three cover variables (above-water cover from riparian vegetation, instream fine cover [e.g. submersed plants], and instream coarse cover [e.g. logs]), as well as a variable identifying the stage of the tidal cycle at the time of sampling. Species richness for both bait trapping

33 and lure fishing was significantly (p <= 0.05) positively correlated with the amount of above-water cover (r = 0.88 and 0.72 respectively). This result corresponds with findings made elsewhere (e.g. Gehrke and Harris [1996] in the Hawkesbury-Nepean River system) which suggest that the integrity of fish communities is positively related to the integrity of the riparian vegetation.

The only other significant correlation was that detected between bait-trapping species richness and the stage of the tidal cycle. The correlation was negative (-0.82) indicating that more species were captured towards low tide. This result is likely to relate to the increased trapping efficiency when fish are concentrated into sheltering sites at low tide (i.e. habitat volume is reduced as tidal waters flow out). From this result it appears important that such sampling be standardised in relation to the tidal cycle, particularly if comparisons are to be made between sites and times. This has a clear implication for future monitoring.

Dataset A2: fish communities from the Newrybar drain and the upper North Creek estuary

This dataset arose from the work of Graham (1989) undertaken in September/October 1989 when a range of netting techniques were used once to samples four sites in the drain and one site in North Creek. The sampling spanned a considerable salinity gradient given salinity differences between sites:

Site 1 (North Ck.): 13.6 ppt Site 2 (drain): 7.6 ppt Site 3 (drain): 3.2 ppt Site 4 (drain): <1.0 ppt Site 5 (drain): <1.0 ppt

Four native primarily-freshwater fish species were recorded in the system. These are listed below together with the highest salinity concentration of water they were recorded from:

firetail gudgeon: <1.0 ppt eeltailed catfish: <1.0 ppt striped gudgeon: 3.2 ppt Australian bass: 7.6 ppt

Dataset B: fish communities in deep sub-tidal habitats in the Richmond River estuary mainchannel

As detailed in Sections 1.6 and 3.1.2, this dataset arose from the work of NSW Fisheries which was initially partially presented by West (1993a&b), then later by West and Walford ( 1999 ) .

Only a suboptimal examination of the fish and salinity data was possible. Changes in fish abundance (% of total numbers caught) along the estuary, and in relation to the available salinity data, are shown in Figures 14a-19a for six species commonly found in freshwater environments.

Two species, the eeltailed catfish (Fig. 19a) and freshwater mullet (Fig. 16a), appeared to be restricted to the upper

34 FIG.14a CHANGES IN THE (%) ABUNDANCE OF AUSTRALIAN BASS ALONG THE RICHMOND RIVER ESTUARY AND IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

Distance (km) from estuary mouth

- Max-sal ------Ave-sal Min-sal MACNO

From West and Walford (1999); sampling in deep mainchannel waters from September 1988 to September 1990.

FIG. 14b CHANGES IN THE (%) ABUNDANCE OF AUSTRALIAN BASS ALONG THE CLARENCE RIVER ESTUARY & IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

— Max-sal ------Ava-sal ~Min-sal ° MACNO

From W63t and Walford (1999); sampling in deep mainchannel waters from September 1988 to September 1990. FIG. 15a CHANGES IN THE (%) ABUNDANCE OF BULLY MULLET ALONG THE RICHMOND RIVER ESTUARY AND IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

Distance (km) from estuary mouth

Max-sal — Ave-sal Min-sal ~ B ~ MUGCE

From W9St and Wallord (1999); sampling in deep mainchannel waters from September 1988 to September 1990.

FIG 15b CHANGES IN THE (%) ABUNDANCE OF BULLY MULLET ALONG THE CLARENCE RIVER ESTUARY & IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

- M ax-sal ------Ave-sal — Min-sal —B_ MUGCE

From West and Wallord (1099); sarrpling In deep mainchannel waters Irom September1988 to September1990. FIG 16a CHANGES IN THE (%) ABUNDANCE OF FRESHWATER MULLET ALONG THE RICHMOND RIVER ESTUARY & IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

Distance (km) from estuary mouth

~ Max-sal ------Ave-sal ~~~ Min-sal MYXPE

From West and WaMord (1999); sampling in deep mainchannel waters from September 1988 to September 1990.

FIG 16b CHANGES IN THE (%) ABUNDANCE OF FRESHWATER MULLET ALONG THE CLARENCE RIVER ESTUARY & IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

— Max-sal •***• Ave-sal —— Min-sal — MYXPE

From Wfcst and Vsfelford (1999); samplins In deep mainchannel waters from September 1988 to September 1990. FIG 17a CHANGES IN THE (%) ABUNDANCE OF FLATHEAD GUDGEON ALONG THE RICHMOND RIVER ESTUARY & IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

Max-sal ------Ave-sal — Min-sal PHIGR

From West and Waltord (1999); sampling in deep mainchannel waters Irom September 1988 to September 1990.

FIG. 17b CHANGES IN THE (%) ABUNDANCE OF FLATHEAD GUDGEON ALONG THE CLARENCE RIVER ESTUARY & IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

— Max-sal —'— Avo-sal Min-sal “e— PHIQR From Woat and Waltord (1000); sampling In deep mainchannel waters from September 1988 to September 1990. FIG 18a CHANGES IN THE (%) ABUNDANCE OF BULLROUT ALONG THE RICHMOND RIVER ESTUARY & IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

Distance (km) from estuary mouth

~*~ Max-sal ------A/e-sal Min-sal ~ B ~ NOTRO

From West and Wallord (1999); sampling In deep mainchannel waters from September 1988 to September 1990.

FIG. 18b CHANGES IN THE (%) ABUNDANCE OF BULLROUT ALONG THE CLARENCE RIVER ESTUARY & IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

— Max-sal *“ “ A/e-sal ‘ Min-sal _s_ NOTRO

From end W»H<«I (1000); sempling In deep mainchannel waters irom September 1988 to September 1990. FIG. 19a CHANGES IN THE (%) ABUNDANCE OF EEL-TAILED CATFISH ALONG THE RICHMOND RIVER ESTUARY & IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

Max-sal ------Ave-sal Min-sal ~ a_ TANSP

From West and Wallord (1999); sampling in deep mainchannel waters from September 1988 to September 1990.

FIG. 19b CHANGES IN THE (%) ABUNDANCE OF EELTAILED CATFISH ALONG THE CLARENCE RIVER ESTUARY & IN RELATION TO SALINITY.

Salinity (ppt) and % of total abundance

■*” Max-sal ------Ave-sal ~~~ Min-sal - s- TANSP

From West and Wallord (1009); sampling In deep mainchannel waters from September 1988 to September 1990. freshwater reaches. The next most restricted species were the bully mullet (Fig. 15a) and flathead gudgeon (Fig. 17a). A very small proportion of the catches of these species arose from the brackish and marine sections of the estuary. A noticeable proportion of Australian bass (Fig. 14a) occurred down to a point in the estuary ( 33 km upstream of the estuary mouth) where salinities reached 5 ppt. Bullrout (Fig. 18a) appeared to be least influenced by salinity being collected in reasonably high proportions in the brackish zone (ending, according to West and Walford [1999], with salinities up to 23 ppt), but less so in the marine zone.

Dataset C: fish communities in deep sub-tidal habitats in the Clarence River estuary mainchannel

This dataset arises from the work of NSW Fisheries which was initially partially presented by West (1993a&b), then later by West and Walford (1999). The work was undertaken in parallel to the investigations in the Richmond River estuary. It involved taking replicated, quarterly samples by trawling over a two-year period along the subtidal environment in the main deepwater channel of the Clarence River estuary. In all 12 systematically- selected longitudinal sites were examined ranging from 5 to 76 km upstream of the estuary mouth (i.e. from upstream of Yamba to upstream of Grafton).

Only a suboptimal examination of the fish and salinity data was possible. Changes in fish abundance (% of total numbers caught) along the estuary, and in relation to the available salinity data, are shown in Figures 14b to 19b for six species commonly found in freshwater environments.

Two species, the eeltailed catfish (Fig. 19b) and freshwater mullet (Fig. 16b), appeared to be most restricted to the upper freshwater reaches. However, in comparison with the Richmond River estuary results, these species were found closer to saline waters. Notably, the catfish was found at a site (~47 km upstream) in which waters reached 1.1 ppt and averaged 0.2 ppt. It is not known whether catfish were actually captured in such salinities as I could not obtain access to data from individual samples.

The next most restricted species were the Australian bass (Fig. 14b). A small proportion of the catches of this species arose from the saline-influenced site mentioned above in relation to catfish. The bully mullet (Fig. 15b) had noticeable proportions of the catch further downstream than the bass.

The bullrout (Fig. 18b), and particularly the flathead gudgeon (Fig. 17b), appeared to be least influenced by salinity being collected in reasonably high proportions in the brackish zone (with salinities up to 16 ppt), but less so in the marine zone.

Dataset D: fish communities in shallow habitats in the Clarence River estuary

This dataset arose from the work of NSW Fisheries which was initially partially presented by West (1993a&b), then later by West and King (1996). It involved the collection of fish by small

35 -mesh seine in three pairs of sites along the estuary every 2 months from September 1989 to July 1990. Four replicates were collected at each site so the total number of samples was 144 (4 replicates x 6 sites x 6 samples). The samples pairs, vegetated and unvegetated, were located in zones referred to as 'freshwater' (~70 km upstream of estuary mouth, near Grafton), 'brackish' (~32 km upstream, near Maclean) and 'marine' (~5 km upstream, near Yamba).

Only a suboptimal examination of the fish and salinity data was possible. Changes in fish abundance (% of total numbers caught) along the estuary, and in relation to the available salinity data, is shown in Figures 20 and 21 for six species commonly found in freshwater environments. The empire gudgeon ('HYPCO' on Figure 20) showed the most pronounced reductions in abundance down the estuary, particularly moving from the freshwater to the brackish zone. The next-most responsive species were the striped and flathead gudgeons ('GOAUS' and 'PHIGR' on Figure 20) which showed major reductions from the brackish zone to the marine zone. The bully mullet and Gambusia ('MUGCE' and 'GAMHO' on Figure 21) did not show a consistent response along the estuary, although both had lowest abundances in the brackish zone. The bullrout ('NOTRO' on Figure 21) increased in abundance going down the estuary, with the largest increase between the brackish and marine zones.

Dataset E: fish communities in tributaries of the lower Clarence River estuary

This dataset arose from the work of NSW Fisheries which was initially partially presented by Pollard and Hannan (1993), then later by Pollard and Hannan (1994). It involved the collection of fish by two methods, poisoning with rotenone and gillnetting, at 18 tributary sites every 3 months from August 1988 to November 1990. Two to three replicate rotenone samples and two replicate gillnet samples were taken per site.

An optimal examination of the fish and salinity data was possible. Only bottom salinity data were utilised. Additionally, only sample data from ungated natural tributaries (4 sites: FN, EN, MN and CN) were examined. This restriction was imposed so to minimise confounding factors arising from the existence of movement barriers (i.e. flood-mitigation gates) on the tributaries.

The cumulative % catches of six key fish species caught in poison stations are shown in Figure 22 in relation to increasing salinity. The cumulative % number of samples in relation to increasing salinity is also shown on this figure (see 'Unique') to provide a frame of reference for the curves depicted for the different fish species. A species which shows no response to salinity would have a curve which approximately follows the 'Unique' curve. The southern blue-eye ('PSESI' on Figure) is an example of this. Species which have curves either starting above, or bending upwards away from the 'Unique' curve, are examples of species showing a negative response to increasing salinity. All

3 6 LL LLI S_ O O L 111 jj o S 1 5 SS ° O 8u j I- 5 m i < » - o c£ 2 6 o _ l LU UJ X < X t- coUJ z—

o *** 0 UJ 111 £ > - CO X 0 < < X CO X s> o u_ UJ o HI X CM UJ « UJ X 0X > LL I- X From From West and King (1996); sampling of shallow habitats from September June 1990. 1989 to o oo

LL LLI S oof o NOTRO

LLI 1X1 2 * 0 dc :

z < o s? CD m i < » - O lO — GAMHO — o ^ z r : o LLI - J 1 < O l— 0 ) Z UJ

<0 % o UJ ol CO o S

< X X CO o CVJ O LL Distance (km) Distance estuary from (km) mouth

OJ o < D LL Max-sal “ — Min sal —— MUGCE

o ------hallow habitats from September 1989 to June 1990. s From From West and King (1996); sampling of FIG. 22 CUMULATIVE (%) CATCHES OF SIX FISH SPECIES IN RELATION TO SALINITY: POLLARD & HANNAN (1994), POISON STNS.

Cumulative % of total

— Unique - + - GAMHO GOBAU HYPCO * - HYPGA — PHIGR PSESI

See text tor species codes. Natural sites FN,EN,MN & CN only in the Clarence River lower estuary. the remaining five species fall into this category. They are in order of apparent in tolerance:

* Gambusia ('GAMHO'): indicative tolerance limit of 0 ppt

* firetail gudgeon ('HYPGA'): indicative tolerance of “1 ppt

* striped gudgeon ('GOBAU'): indicative tolerance of “7 ppt (i.e. its curve commences to bend upwards from the 'Unique' curve at this point)

* empire gudgeon ('HYPCO'): indicative tolerance of ~11 ppt

* flathead gudgeon ('PHIGR'): indicative tolerance of ~11 ppt

The cumulative % catches of five key fish species caught in gillnetting stations are shown in Figure 23 in relation to increasing salinity. As before, the cumulative % number of samples in relation to increasing salinity is also shown on this figure (see 'Unique'). All the species either have curves starting above, or bending upwards away from the 'Unique' curve. They therefore show an apparent negative response to increasing salinity. Their apparent tolerance limits are:

* bullrout ('NOTRO'): indicative tolerance limit of 0 ppt

* freshwater herring ('POTRI'): indicative tol. limit of 0 ppt

* freshwater mullet ('MYXPE'): indicative tol. limit of 0 ppt

* forktailed catfish ('ARIGR'): indicative tolerance of “1 ppt

* Australian bass ('MACNOV'): indicative tolerance of ~3 ppt

Dataset F: fish communities in the Hawkesbury-Nepean River estuary

This dataset arises from the work of NSW Fisheries which was partially presented by Pollard e t a l. (1994) and Gehrke and Harris (1996). The dataset can be divided into the following five subsets:

Electrofishing I: Electrofishing in the littoral zone of the estuary at nine sites spread from near the tidal limit to a point ~70 km downstream (which is “80 km upstream from the estuary mouth). Four replicates were taken at each site in four seasons (late-'92 to early-sping'93) resulting in the collection of 144 samples.

Electrofishing II: Electrofishing in the littoral zone of the estuary at two sites (one grassed, one well-vegetated), one close to the tidal limit and the other “50 km downstream. Four replicates were taken at each site in four seasons (December'93 to September'94) resulting in the collection of 32 samples.

Electrofishing III: Electrofishing in the littoral zone of the estuary at eight sites (+- nutrient enriched, +-degraded banks),

37 FIG. 23 CUMULATIVE (%) CATCHES OF FIVE FISH SPECIES IN RELATION TO SALINITY: POLLARD & HANNAN (1994), GILLNET STNS.

Cumulative % of total

----- Unique - + - POTRI MYXPE - Q- NOTRO MACNOV ARIGR

See text tor species codes. Natural sites FN,EN,MN & CN only in the Clarence River lower estuary. spread from near the tidal limit to a point ~A0 km downstream. Four replicates were taken at each site in four seasons (December'93 to September'94) resulting in the collection of 128 samples.

Gillnetting I: Gillnetting with various mesh-sized nets in the littoral zone of the estuary at two sites, one close to the tidal limit and the other ~70 km downstream. One sample was taken at each site in four seasons (winter'85 to autumn'86) resulting in the collection of 8 samples.

Gillnetting II: Gillnetting with various mesh-sized nets in the littoral zone of the estuary at two sites, one close to the tidal limit and the other ~70 km downstream. One sample was taken at each site in four seasons (summer'94 to spring'95) resulting in the collection of 8 samples.

Dr P. Gerkhe indicated that while the fish-sample data is located on NSW Fisheries' databases, there is only very limited parallel- collected salinity data present. This appears to be a result of either the adhoc collection of salinity data, or the adhoc addition of the data onto the databases. Accordingly, this data could not be utilised in the present study to develop indicative salinity criteria. There remains a possibility, at least for the electrofishing datasets, that parallel salinty data are present on raw data forms. In using an electrofisher it is necessary to measure the electrical conductivity of water to set voltages which allow effective fishing. These are usually recorded on a data form for each operation.

38 4.0 DISCUSSION AND CONCLUSIONS

4.1 High value components

High-value components of fish habitats and the fauna were identified in Section 3.2. A schematic summary of their distribution is given in Figure 24. The upper arms of the estuary have been focussed on as these primarily freshwater environments are most vulnerable to shifts in salinity structure. It is emphasised that the identification process was done on a desktop-analysis basis and is therefore preliminary. Nevertheless, following the precautionary principle, all efforts should be made to minimise impacts on the identified components.

Fish habitats. There are clear indications that the Bungawalbin Creek arm of the estuary potentially contains extensive high- value physical habitat. Unfortunately, there are also some indications that the value of this habitat has recently been reduced by deteriorating water quality.

To obtain more definition and confidence in the identification of high-value habitats, habitat surveys along the upper arms of the estuary would need to be undertaken. They would need to focus on the quantifcation of:

* above-water cover provided by riparian vegetation, * instream-fine cover provided by, for example, aquatic plants * instream-coarse cover provided by, for example, submersed logs * habitat diversity

Fish fauna. The upper Richmond River arm of the estuary potentially has the greatest complement of high-value fauna given the additional records of the high-conservation-value Oxleyan pygmy perch and Eastern freshwater cod. These records need to be confirmed and this would only be possible through intensive, well-focussed sampling. It would be advisable to undertake such sampling in the other upper arms of the estuary to verify the apparent uniqueness of the upper Richmond arm.

4.2 Inconsistencies between derived indicative salinity tolerances and the associated complexities

There are many inconsistencies between indicative salinity tolerances derived from the literature/discussions (Section 3.3.1) and those derived from the analysis of available data along salinity gradients (Section 3.3.2). There are also considerable inconsistencies amongst the indicative tolerances derived from the salinity-gradient data. These inconsistencies are summarised in Table 6.

The basis of the inconsistencies is not clear. Many explanations are possible. They may relate to the narrow spatial focus of the field investigations, the fact that a suboptimal examination of the salinity-gradient data was only possible in most cases, or as mentioned earlier, absence or reduced abundance may be due to factors other than salinity tolerance. Additionally, there may be

39 FIG. 24 SCHEMATIC DISTRIBUTION OF IDENTIFIED HIGH-VALUE COMPONENTS OF THE RICHMOND RIVER ESTUARY.

Habitat comoonanti ara ahovn for only kay faaturaa:

a ■ hlghaat atraaa haalth ratine (SHR) ra riparian vagatatlon condition; b ■ hlghaat SHR ra aquatic C ■ hlghaat SHR ra ovarall condition; d ■ hlghaat riparian vagatatlon covar (RVC) for > 25 i vida catagory; 1 ■ hlghaat RVC for > SO ■ catagory; t ■ hlghaat RVC for > 100a catagory; g • hlghaat channal atructura EFC dlvaraity lndax (CSOI); h ■ cloaa aacond hlghaat CSD1; 1 ■ hlghaat fraquancy of channal bad anoaallaa; 1 ■ cloaa aacond hlghaat RVC for > 23 a catagory.

ElfJl__fauna coaoonanta ara alao ahovn for only kay faaturaa;

EfC ■ laatarn fraahvatar cod; OFF ■ Oxlayan pygmy parch OPP -* • Tha othar high-valua apaclaa (••ltailad catflah, Auatrallan baaa, bully nullat and fraahvatar mullat) ara llkaly to occur throughout tha uppar arma of tha aatuary.

g

Casino"

/ \ w Q Q / ^Woodburn Q O •Ballina ** U M dS Pi w o> _] 1 0 km SEA TABLE 6 INCONSISTENCIES BETWEEN DERIVED INDICATIVE SALINITY TOLERANCES. Tolerances derived from the literature and discussions are coded L/D. Tolerances derived from salinity gradient data are coded SGD. The dataset coding is indicated in Section 3.3.2.

Literature/discussions (L/D) vs. salinity gradient data (SGD)

High-value components of the fauna:

* freshwater mullet: L/D indicates that they move to the sea, yet SGD (datasets B and C) indicates that they are restricted to the upper freshwater zone of the estuary * bully mullet: L/D indicates that they move to the sea, yet SGD indicates there is some restriction to the upper freshwater zone (datasets B and C), or is not responsive to differences in salinity (dataset D) * Australian bass: L/D indicates 5 ppt salinity upper limit outside of the spawning season, yet three SGD datasets (A2, C and E) indicate 7.6, 1 and 3 ppt respectively for upper limits (September/October for the first, no season specified for the latter two)

Non-high-value components of the fauna:

* bullrout: L/D indicates that the species is catadromous, yet SGD for dataset E suggests that there is no tolerance of saline waters * Gambusia: indicates that the species can tolerate full marine salinities, yet SGD for dataset E suggests there is no tolerance of saline waters * forktailed catfish: indicates that the species can tolerate full marine salinities, yet SGD for dataset E suggests the upper tolerance may be 1 ppt

Amongst salinity-gradient datasets

High-value components of the fauna:

* bully mullet: datasets B and C indicates there is some restriction to the upper freshwater zone, yet dataset D indicates that it is not responsive to differences in salinty * Australian bass: dataset B indicates 5 ppt salinity upper limit, yet datasets E indicates 3 ppt, and further, dataset C indicates 1 ppt; dataset A2 gave a maximum value 7.6 ppt during the non-spawning season (September/October)

Non-high-value components of the fauna:

* bullrout: dataset E indicates no tolerance to saline waters, yet datasets B and C indicate weak restrictions to the freshwater zone, and further, dataset D indicates they prefer higher salinities considerble differences in tolerances between lifecycle stages and this was not examined in the present investigation. Furthermore, between-catchment differences in tolerance may exist for a single species (Hart e t al. 1991).

The inconsistencies may also arise from the inherent weaknesses of salinity-tolerance data derived from laboratory work. Williams and Williams (1991) stated that extreme caution should be used when extrapolating laboratory-derived tolerance data as the range of life-history stages may show considerable differences. Hart et al. (1991) indicated that a short-term laboratory test does not represent the ability of a population to survive and reproduce indefinitely under a particular salinity regime.

Hart el a l. (1991) indicated that behavioural responses to salinity change have been poorly studied. They quoted an example (Bulkley [1983], details of the reference not given in their paper) where fishes actively preferred habitats with salinities far less than their salinity tolerances, i.e. avoidance behaviour. There is also a dearth of information on how salinity affects competitive and predatory interactions.

It is recognised that virtually all the fish species typically found in freshwater environments in the Richmond River catchment do not have long evolutionary histories in freshwaters. They are derived largely from recent marine ancestors, with the possible exception of the eeltailed catfish (Plotosidae). Accordingly, it could be expected that many would retain at least moderate salinity tolerances, i.e not be strongly stenohaline. With such an understanding there is conceptually a reduced importance of direct physiological impacts arising from increases in salinity (i.e. the physiological aspect of Mechanism FRAIE3 highlighted in Section 1.5). Instead more attention should generally be focussed on i) direct impacts causing adverse behavioural changes (e.g. the navigation aspect of Mechanism FRAIE13), and ii) indirect impacts such as habitat degradation (i.e. Mechanism FRAIE4) , competition and predation arising from species with markedly-high salinity tolerances (i.e. the species interaction aspect of Mechanism FRAIE3). Salinity tolerance may have some relevance, but not necessarily the tolerances of the typically freshwater species, but instead, the lower salinity tolerance of the more- marine species.

Knowing the salinity tolerance of an organism has little meaning unless there is some specification of the duration and frequency of exposure to a particular salinity level. For example, an exposure of 1 minute duration is likely to be less harmful than an exposure of 1 week. Similarly, a 4-day exposure once a year is likely to be less harmful than a 4-day exposure once a week. There is virtually no information available on this aspect of salinity tolerance.

4.3 Using indicative salinity limits

Recommended indicative salinity limits arising from this study are given in Table 7. These are essentially the same as listed in Section 3.3.1. The primary change, as arising from the examination of salinity-gradient data, is the inclusion of the eeltailed catfish ( Tandanus sp.A; a species which potentially has

4 0 TABLE 7 SUMMARY OF INDICATIVE SALINITY LIMITS ARISING FROM THE PRESENT STUDY.

0.12 ppt: upper limit for Oxleyan pygmy perch

1 ppt: upper limit for Eastern freshwater cod (indirect impacts), eeltailed catfish ( Tandanus sp.A) and the maintenance of freshwater ecosystems

5 ppt: upper limit for adult Australian bass outside the spawning season as well as for ribbon grass which is important shelter for larval and juvenile bass; possibly also relevant to the common reed which also provides important shelter.

8-13 ppt: lower and upper limit for adult Australian bass during the spawning season.

20 ppt: indicative limit for the Sydney rock oyster above which there is an increased chance of fouling of attachment substrates by marine animals. a high conservation value) in the group which is likely to have a 1 ppt salinity limit.

The discussion in Section A. 2 illustrates the complexities involved in understanding the way in which salinity influences fish communities in the upper arms of estuaries. It also illustrates the potential weaknesses of any derived salinity limits. Nevertheless, there is a useful role for such limits if they are based on 'best-available' information. Being indicative of points on the salinity spectrum where biological changes may commence, they can be used as 'working' thresholds to be inputed into risk analyses which assess the implications of extracting particular volumes of freshwater. This approach to assessing provisions has generally been accepted in Australia (Arthington 1999).

A useful 'tool' in this regard are graphed relationships between the impactor on the X-axis, e.g. the volume of water extracted, and the impacted on the Y-axis, e.g. the area of critical habitat remaining (e.g. the area of the estuary with salinity below the 'working' threshold [e.g. < 1 ppt]). Of key interest on such relationships are bands where slope increases noticeably. That is, where small increases in extraction result in large reductions in critical habitat. It is clearly important to avoid such slope bands where possible.

The risk-analysis process helps to identify characteristics of extractions (e.g. volume, rate, and timing) which should minimise resultant impacts. Important here is the provision of a means to understand and demonstrate the nature of tradeoffs between increased water extraction, and increased impacts on the estuarine ecosystem. The process as described has the advantage that it attempts to utilise what information is available, or readily attainable, on individual estuaries and their most valuable and/or vulnerable components. Of fundamental importance, the process utilises estuary-specific information on known links between hydrology and basic structural properties of the ecosystem.

A.A Adaptive management and monitoring

Given the complexities of estuarine ecosystems and inaccuracies which may occur in the methodologies, particularly in relation to the range of working thresholds utilised, it is imperative that any implemented extraction regime be viewed as an interim condition, to be revised once substantial knowledge is gained through ensuing scientific research and monitoring. This is a fundamentally important feature of any adaptive management system ( sensu Knights and Fitzgerald 199A).

Monitoring in estuarine ecosystems is potentially more difficult than in river/stream ecosystems due to the complexities introduced by the daily tidal cycle. To enable comparisons between places and times, samples should be taken at the same phase of the tidal cycle (i.e. high, mid-coming-in, mid-going out, or low), as well as the same time of day. Attempts to utilise comparable combinations of tidal phase, and time of day, will result in protracted sampling sessions. The importance of standardising in relation to tidal phase was demonstrated by

A1 Bishop (1999) in a study on fish communities in the upper Emigrant Creek estuary. A significant correlation was found between fish species richness at given sites and the tidal phase when each site was sampled. This result, which indicated that more species were captured towards low tide, was considered to be caused by increased trapping efficiency when fish were concentrated into sheltering sites at low tide (i.e. habitat volume reduces when tidal waters flow out).

5.0 ACKNOWLEDGMENTS

For the provision of valuable information on the Richmond River estuary, and its fauna, I thank the following people from major organisations: Mr A. Moore and Ms K. Toissant (Southern Cross University), Ms S. Fairfull, Dr D. Pollard, Dr B. Pease, Mr G. Gordon, Dr S. Rowland and Mr R. Williams (NSW Fisheries), Mr M. McGrouther (Australian Museum), Mr W. Garrard (Richmond Catchment Management Committee), Dr R. West (Wollongong University) and Mr M. Chadwick (Water Research Laboratory).

For valuable information on recreational fisheries in the estuary I thank the following anglers: Mr G. Dodd, Mr L. Doust, Mr A. Henderson, Mr F. Sherwood and Mr M. Youman.

I additionally acknowledge Ms M. Dalby-Ball (Sainty & Associates) and Dr E. Avery (DLWC, Newcastle) for alerting me to some pertinent publications. Dr T. Grant, Mr B. Peirson and Mr M. Chadwick are thanked for thoughtful discussions on relevant issues.

4 2 6.0 REFERENCES

Allen, L.G. (1982). Seasonal abundance, composition and productivity of the littoral fisf assemblage in upper Newport Bay, California. Fish. Bull. U. S .,80, 769-790.

ANZECC (1992). National water quality management strategy: Australian water quality guidelines for fresh and marine waters. Australian and New Zealand Environment and Conservation Council. November 1992.

Arthington, A .H. (Ed.) ( 1996 ). The recovery plan for the Oxleyan Pygmy Perch, Nannoperca oxleyana. Centre for Catchment and In- stream Research, Griffith University. Final report to the Australian Nature Conservation Agency. February 1996.

Arthington, A.H. (1999). System-wide assessment of opportunities to develop integrated water and sewerage strategies in the Lower Clarence River Region: Independent review of recent studies on the environmental flow requirements in the Nymboida- System. Griffith University, Queensland.

Austin, M.W. (1992). A study of the freshwater herring (Potamalosa richmondia) in the Richmond River with emphasis on age and growth, diet reproductive biology and habitat. Integrated project as a partial fulfilment of the Bachelor of Applied Science (Coastal Management). Southern Cross University.

Battaglene, S.C. and Talbot, R.B. (1993). Initial swim bladder inflation in intensively reared Australian bass larvae, Macquaria novemaculeata (Steindachner)(Perciformes: Percichthyidae). Aquaculture, 86, 431-442.

Bird, L. (1997). The Richmond River catchment stream health assessment report. Department of Land and Water Conservation, November 1994.

Bishop, K.A. (1999a). Nymboida/Timbarra Water Diversion Scheme scoping study. Study component: prominent issues concerning impacts on fish. Report prepared for the Centre for Water Policy Research, University of New England.

Bishop, K.A. (1999b). Revised eight-part tests on three threatened fish species in relation to the Pacific Highway upgrading: the Ballina bypass. Report prepared for Connell Wagner Pty. Ltd. on behalf of the NSW Roads and Traffic Authority. May 1999.

Bishop, K.A. (1999c). Threatened fish species surveying and habitat assessment in the Emigrant Creek catchment, Richmond River system, New South Wales. Report prepared in relation to the Pacific Highway upgrading: the Ballina bypass. Report to Connell Wagner Pty. Ltd. on behalf of the NSW Roads and Traffic Authority. September 1999.

Bishop, K.A. (1995). Hastings district water supply augmentation: freshwater ecology study for the intake pump station upgrading at Koree Island. Report to the NSW Public Works through Connell

43 Wagner Pty. Ltd.

Blaber, S.J.M. (1997). Fish and fisheries of tropical estuaries. Fish and Fisheries Series, 22. Chapman and Hall, London, Great Britain,

Blaber, S.J.M. and Blaber, T.G. (1980). Factors affecting the distribution of juvenile estuarine and inshore fish. Journal o f Fish Biology , 17, 143-162.

Bulger, A.J., Hayden, B.P., Monaco, M.E., Nelson, D.M., and McCormickray, M.G. (1993). Biologically-based estuarine salinity zones derived from multivariate analysis. E stu a rie s, 16, 311-322.

Bunn, S.E., Loneragan, N.R. and Yeates, M. (1998). The influence of river flows on coastal fisheries. In: Arthington, A.H. and Zalucki, J.M. (eds.) Comparative evaluation of environmental flow assessment techniques: Review of methods. Land and Water Resources Research Development Corporation, Occasional Paper No. 27/98 .

Cody, A. (1995). The fish assemblages associated with seagrass habitats in the Richmond River estuary, New South Wales. Integrated project as a partial fulfilment of the Bachelor of Applied Science (Fisheries Management and Aquaculture). Southern Cross University.

Currie, B. (1998). Distribution and abundance of macrofauna in seagrass habitats in the Richmond River estuary, N.S.W. Integrated project as a partial fulfilment of the Bachelor of Applied Science (Environmental Resource Management). Southern Cross University.

Cyrus, D.P. and Blaber, S.J.M. (1987a). The influence of turbidity on juvenile marine fishes in estuaries. Part 1, field studies. Journal of Experimental and Ecology , 109, 53-70.

Cyrus, D.P. and Blaber, S.J.M. (1987b). The influence of turbidity on juvenile marine fishes in estuaries. Part 2, laboratory studies. Journal of Experimental Marine Biology and Ecology , 109, 71-91.

Day, G. (1994). An assessment of the relationships between in- stream habitat and the freshwater fish populations of Bungawalbin Creek, Coraki, northern NSW. Integrated project as a partial fulfilment of the Bachelor of Applied Science (Fisheries Management and Aquaculture). Southern Cross University.

Doohan, M. (1989). Native versus introduced. A plan of management for native and introduced fish for the Richmond River and its tributaries. Integrated project prepared as partial requirements of the Bachelor of Applied Science (Coastal management), University of New England, Northern Rivers Campus.

Drinkwater, K.F. and Frank, K.T. (1994). Effects of river regulation and diversion on marine fish and invertebrates. Aquatic Conservation: freshwater and marine ecosystems, 4, 135- 151. Dunk, C. (1991). An evaluation of the spiny , Euastacus v a le n tu lu s, in the Richmond River drainage. Integrated project prepared as partial requirements of the Bachelor of Applied Science (Coastal management), University of New England, Northern Rivers Campus.

Dwyer, P. (1996). Variability of intertidal benthic macro­ communities in sandy sediments along a gradient of wave exposure. Integrated project as a partial fulfilment of the Bachelor of Applied Science (Coastal Management). Southern Cross University.

Earley, 0. (1999). Richmond River Catchment Wetlands Inventory. Richmond Catchment Management Committee.

Farley, C.A., Wolf, P.H. and Elston, R.A. (1988). A long-term study of 'microcell' diseases in oysters with a description of a new genus -Mikrocytos (G.N.) and two new species M ikrocytos mackini (sp. n.) and Mikrocytos roughleyi (sp. n.). Fish. Bull., 83, 581-593.

Gehrke, P.C., and Harris, J.H. (Eds.)(1996). Fish and fisheries of the Hawkesbury-Nepean River System. Final report to the Sydney Water Corporation. Co-operative Research Centre for Freshwater Ecology and NSW Fisheries Research Institute, May 1996.

Glaister, J.P. (1978). The impact of river discharge on distribution and production of the school prawn Metapenaeus macleayi (Haswell)(Crustacea: Penaeidae) in the Clarence River region, northern New South Wales. Aust. J. Mar. Freshwat. Res., 29, 311-323.

Graham, M.K.B. (1989). Fish utilisation of the Newrybar flood mitigation drain and fish habitat management of flood mitigation schemes. Thesis, University of New England (Northern Rivers), Lismore, New South Wales.

Gray, C.A., McDonall, V.C. and Reid, D.D. (1990). By-catch from prawn trawling in the , New South Wales: species composition, distribution and abundance. Aust. J. Mar. Freshwater R es., 41, 13-26.

Growns, I.O., Pollard, D.A. and Gehrke, P.C. (1998). Changes in river fish assemblages associated with vegetated and degraded banks, upstream of and within nutrient-enriched zones. Fish. Man. & E c o l., 5, 55-69.

Harris, J.H. (1983). The Australian bass, Macquaria novemaculeata. Ph.D. Thesis, University of New South Wales.

Harris, J.H. (1985). Diet of the Australian bass, Macquaria novemaculeata, (Perciformes, Percichthyidae) in the . Aust. J. Mar. Freshwater Res., 36, 219-34.

Harris, J.H. (1986). Reproduction of the Australian bass, Macquaria novemaculeata, (Perciformes, Percichthyidae) in the Sydney Basin. Aust. J. Mar. Freshwater Res., 37, 209-35.

45 Harris, J.H. (1988). Demography of the Australian bass, Macquaria novemaculeata, (Perciformes, Percichthyidae) in the Sydney Basin. Aust. J. Mar. Freshwater Res., 39, 355-69.

Harris, J.H. and Gehrke, P.C. (Eds.)(1997). Fish and rivers in stress. The NSW Rivers Survey. Report published by NSW Fisheries Office of Conservation and the Cooperative Research Centre for Freshwater Ecology in association with NSW Resource and Conservation Assessment Council.

Hart, B.T., Bailey, P., Edwards, R., Hortle, K., James, K., McMahon, A., Meredith, C. and Swadling, K. (1991). A review of salt sensitivity of the Australian freshwater biota. Hydrobiologia, 210, 105-144.

Hoff, J.G. and Ibara, R.M. (1977). Factors affecting the seasonal abundance, composition and diversity of fishes in a southeastern New England estuary. Estuar. c s tl mar. Sci, 5, 665-678.

Holliday, F.G.T. (1971). Salinity: Fishes. In: Marine ecology, Vol.2. pp997-1033. Ed. by 0. Kinne. New York: Wiley-Interscience.

Howland, M.B. (1998). Clarence / Coffs Harbour Regional Water Supply Project. Environmental Assessment - Working Paper No.9. Estuarine flora and fauna. Report prepared for the Department of Public Works and Services, NSW. Southern Cross University, final draft, July 1998.

Jerry, D.R. and Woodland, D.J. (1997). Electrophoretic evidence for the presence of the undescribed 'Bellinger' catfish (Tandanus sp.) (Teleostei: Plotosidae) in four New South Wales mid-northern coastal rivers. Mar. Freshwater Res., 48, 235-40.

Kinne, 0. (1964). The effects of temperature and salinity on marine and brackish water animals. Oceanogr. mar. B io l. A. Rev., 2, 281-339.

Knights, P. and Fitzgerald, B. (1994). A pragmatic approach to environmental flow management. Proceedings of the Environmental Flows Seminar, Australian Waste Water Association Incorporated, Artarmon, Australia.

Koehn, J.D. and O'Conner (1990). Biological information for management of native freshwater fish in Victoria. Victorian Government Printing Office on behalf of the Freshwater Fish Management Branch, Dept. Cons, and Envirn., Melbourne.

Liggins, G.W., and Kennelly, S.J. (1996). By-catch from prawn trawling in the Clarence River estuary, New South Wales, Australia. Fisheries Research, 25, 347-367.

Loneragan, N.R., Potter, I.C., Lenanton, R.C.J. and Caputi, N. (1986) . Spatial and seasonal differences in the fish fauna in the shallows of a large Australian estuary. Marine Biology, 92, 575- 586 .

Loneragan, N.R., Potter, I.C., Lenanton, R.C.J. and Caputi, N. (1987) . Influence of environmental variables on th fish fauna of th deeper waters of a large Australian estuary. Marine Biology,

46 94, 631-641.

McDowall, R.M. (Ed.)(1996). Freshwater fishes of South-eastern A u stra lia . Reed Books, Sydney.

Marchand, J. (1993). The influence of seasonal salinity and turbidty maximum variations on the nursery function of the Loire estuary (France). Netherlands Journal of Aquatic Ecology, 27, 426-436.

Marias, J.F.K. (1988). Some factors that influence fish abundance in South African estuaries. Estuarine, Coastal and Shelf Science, 6, 67-77.

Marshall, S. and Elliot, M. (1998). Environmental influences on the fish assemblage of Humber Estuary, U.K. Estuarine, Coastal and Shelf Science,46, 175-184.

Melville, J.C. (1991). A review of the biology and ecology of the freshwater crayfish, Euastacus valentulus and their distribution within Emigrant Creek, northern New South Wales. Integrated project prepared as partial requirements of the Bachelor of Applied Science (Coastal management), University of New England, Northern Rivers Campus.

Merrick, J.R. and Schmida, E.G. (1984). Australian freshwater fis h e s . Griffin Press: Netley, South Australia.

MHL (1998). Freshwater extraction from the Richmond River below tidal limit. Manly Hydraulics Laboratory Report No. 937 prepared for the Department of Land and Water Conservation.

Nell, J.A. and Holliday, J.E. (1988). Effects of salinity on the growth and survival of Sydney rock oyster (Saccostrea commercialis) and Pacific oyster (Crassostrea gigas) larvae and spat. Aquaculture, 68, 39-44.

Nihill, K.J. (1994). An assessment of the recreational fishery based on the Richmond River-estuary, Ballina N.S.W. Integrated project as a partial fulfilment of the Bachelor of Applied Science (Fisheries Management and Aquaculture). Southern Cross University.

Odum, W.E (1970). Insidious alteration of the estuarine environment. Trans. Am. Fish Soc., 88, 836-847.

Pearce, A. (1994). Testing the relative selectivity of various methods applicable for surveying fish in freshwater streams. Integrated project as a partial fulfilment of the Bachelor of Applied Science (Fisheries Management and Aquaculture). Southern Cross University.

Pease, B.C. and Scribner, E.A. (1993). New South Wales commercial fisheries statistics 1990/91. NSW Fisheries, Fisheries Research Institute, Cronulla.

Pease, B.C. and Grinberg, A. (1995). New South Wales commercial fisheries statistics 1940-1992. NSW Fisheries, Fisheries Research Institute, Cronulla.

47 Peterson, M.S. and Ross, S.T. (1991). Dynamics of littoral fishes and decapods along a coastal river-estuarine gradient. E stu a rin e, Coastal and Shelf Science,33, 467-483.

Peverell, S. (1996). Finfish database for the lower Richmond River estuary, Ballina NSW. Integrated project as a partial fulfilment of the Bachelor of Applied Science (Fisheries Management and Aquaculture). Southern Cross University.

Pollard, D.A. and Hannan, J.C. (1993). Nothern Rivers Report Part B: effects of structural flood mitigation works on fish communities. NSW Fisheries Research Internal Report, December 1993 .

Pollard, D.A. and Hannan, J.C. (1994). The effects of structural flood mitigation works on fish habitats and fish communities in the lower Clarence River system of south-eastern Australia. E stu a ries, 17, 427-461.

Pollard, D.A., Growns, 1.0, Pethebridge, R.L. and Marsden, T.J. (1994). Hawkwsbury-Nepean fish ecology study. Six-monthly report, May 1994. Report by NSW Fisheries Research Institute, Cronulla, to the Sydney Water Board.

Potter, I.C., Claridge, P.N., Warwick, R.M. (1986). Consistency of seasonal changes in an estuarine fish assemblage. Mar. E c o l. Prog. Ser., 32, 217-228.

Quinn, N.J. (1980). Analysis of temporal changes in fush assemblages in fish assemblages in Serpentine Creek, Queensland. Envir. Biol. Fish., 5, 117-133.

Reid, D. (1992). Benthic macrofauna associated with seagrass and sand habitats in the Richmond River estuary, Ballina, N.S.W. Integrated project as a partial fulfilment of the Bachelor of Applied Science (Coastal Management). Southern Cross University.

Rochford, D.J. (1951). Studies of Australian estuarine hydrology. I. Introductory and comparative features. Aust. J. Mar. Freshwat. R es., 2, 1-116.

Roughley, T.C. (1926). An investigation of the cause of an oyster mortality on the Georges’ River, New South Wales. Proc. Linn. Soc. NSW., 51, 446-491.

Ruello, N.V. (1973). The influence of rainfall on the distribution and abundance of the school prawn Metapenaeus macleayi in the Hunter region. Marine Biology, 23, 221-228.

Sainty, G.R. and Jacobs, S.W.L. (1981). Waterplants of New South Wales. New South Wales Water Resources Commission, Sydney.

Sainty, G. and Jacobs, S. (1999). Emigrant Creek macrophytes and riparian vegetation - assessment of ’health'. Report to Connell Wagner Pty. Ltd. on behalf of the NSW Roads and Traffic Authority, July 1999.

Sheaves, M. (1998). Spatial patterns in estuarine fish in

4 8 tropical Queensland: a reflection of interaction between long­ term physical and biological processes? Marine and Freshwater Research, 49, 31-40.

Stewart, C. (1998). A freshwater fish species inventory of three lagoons in the Bungawalbin Catchment, northern NSW. Integrated project prepared as partial fulfilment of the requirements of the Bachelor of Applied Science ( and Management), Southern Cross University.

Summers, D. (1997). Review of the decline of seagrass habitats in the lower Richmond River using remote sensing techniques. Integrated project as a partial fulfilment of the Bachelor of Applied Science (Fisheries Management and Aquaculture). Southern Cross University.

The Ecology Lab Pty. Ltd. (1996). Shoalhaven water supply augmentation EIS. Aquatic ecology and fisheries downstream of B urrier. Report to Dames and Moore, October 1996.

Theil, R.A., Sepulveda, A., Kaffman, R. and Nellen, W. (1995). Environmental factors structuring the fish community of the Elbe estuary. Journal of Fish Biology, 46, 47-69.

Van der Wal, E.J. (1985). Effects of temperature and salinity on the hatch rate and survival of Australian bass (Macquaria novemaculeata) eggs and yolk-sac larvae. Aquaculture, 47, 239- 244 .

Weinstein, M.P., Weiss, S.L and Walters, M.F. (1980). Mulitple determinants of community structure in shallow marsh habitats, Cape Fear River estuary, North Carolina, USA. Marine Biology, 58, 227-243.

West, R.J. and Walford, T.R. (1999). Species distributions, biomass and structuring of fish communities along the main channels of two large eastern Australian estuaries. Submitted to E stu a rie s.

West, R.J. (1993a). Estuarine fisheries resources of two south­ eastern Australian Rivers. PhD thesis, University of NSW.

West, R.J. (1993b). Northern Rivers Report Part A: Estuarine Fisheries Resources. NSW Fisheries Research Internal Report, December 1993.

West, R.J. and Gordon, G.N.G. (1994). Commercial and recreational harvest of fish from two Australian Coastal rivers. Aust. J. Mar. Freshwater Res., 45, 1259-79.

West, R.J., Thorogood, C.A., Walford, T.R. and Williams, R.J. (1985). An estuarine inventory for New South Wales, Australia. Fisheries Bulletin 2, NSW Agriculture & Fisheries, Sydney, Australia.

West, R.J. and King, R.J. (1996). Marine, brackish, and freshwater fish communities in the vegetated and bare shallows of an Australian coastal river. E stuaries, 19, 31-41.

4 9 Whitfield, A.K. (1996). Fishes and the environmental status of South African estuaries. Fisheries Management and Ecology, 3, 45- 57 .

Whitfield, A.K., Blaber, S.J.M., and Cyrus, D.P. (1981). Salinity ranges of some south African fish species occurring in estuaries. S. Afr. J. Zool., 16, 151-155.

Williams, M.D. and Williams, W.D. (1991). Salinity tolerances of four species of fish from the Murray-Darling River system. Hydrbiologia , 210, 145-160.

Wolf, P.H. (1967). Winter mortality hits state's oyster output. The Fisherman, June 1967.

50 A ppendix C

Distribution of Platypuses and Platypus Habitat in the Richmond River Estuary and Investigation Salinity Tolerance in the Species

for Dr K A Bishop Freshwater Consultant

by

T R Grant Education and Environment Services Pty Ltd 2

Abstract

Data gathered from literature and data base searches, including information from the Richmond River estuary, indicated that platypuses are largely restricted in their distribution to the upper sections of estuaries, within a few kilometres of the limit of tidal influence. These sections of the Richmond River estuary are essentially fresh water (< 0.5 ppt). The reasons for such distribution is not known, but is probably related to a number and!or combination of biotic and abiotic factors influenced by tidal cycles in the estuarine environment, rather than to the species’ tolerance to a specific level of a single factor, such as salinity. Encroachment of greater salinity further up the arms of the Richmond River estuary, due to the extraction of water from tidal pools, may not directly affect the platypus in terms of its tolerance of higher saline conditions, but may displace whatever factors are determining the distribution of the species further upstream, thereby reducing its distribution within the river system

Salinity changes, brought about by water extractions from the various tidal pools in the Richmond River estuary system, could affect a number of biotic and abiotic factors influencing platypus distribution in the estuarine part of the river system and it is recommended that

• any proposed extraction regimes be modelled as accurately as possible.

• an adaptive management strategy be developed, which includes strict monitoring programs and a facility for change to extraction regimes in response to either environmentally positive or negative findings of monitoring studies.

• at least initially no extraction, or very limited extraction, be made from the smaller ttdal pools, including Bungawalbin and Emigrant Creeks.

Appendix C - Platypus 3 TABLE OF CONTENTS

Abstract 2

1.0 INTRODUCTION 4 1.1 Background to the study 1.2 Distribution of platypuses in relation to salinity 2.0 METHODS 5 2.1 Platypus distribution in estuaries 2.2 Determination of salinities in estuaries 2.3 Investigation of other habitat variables 3.0 RESULTS 5 3.1 Occurrence of platypuses in estuaries - General 3.2 Occurrence of platypuses in estuaries - New south Wales 3.3 Occurrence of platypuses in estuaries - Richmond River 3.4 Matching salinity data to platypus records in the Richmond River estuary 3.4.1 Emigrant Creek 3.4.2 Maguires Creek 3.4.3 Tucki Tucki Creek 3.4.4 Bungawalbin Creek 3.4.5 Richmond River 3.4.6 Wilsons river 3.4.7 Leycester, Terania and Goolmangar Creeks 3.5 Habitat assessment

4.0 DISCUSSION 8 4.1 Distribution of platypuses and platypus habitat in estuaries 4.2 Determinants of platypus distribution in estuaries 4.3 Consequences of the observed estuarine distribution of platypuses 4.4 Limitations of the study

5.0 CONCLUSIONS 10

6.0 BIBLIOGRAPHY 11 LIST OF TABLES

1.1 Salinity guidelines for freshwaters in Australia 13 1.2 Platypus occurrence in streams of “Poor” or very “Fair” salinity conditions in New South Wales 13

3.1 Reported occurrences of platypuses in estuaries of New South Wales 14

3.2 Platypus records around and downstream of the tidal limits in the Richmond River and its tributaries 15

Appendix C - Platypus 4 1.0 INTRODUCTION

1.1 Background to study

The water in tidal pools has been used to provide fresh water supplies during low river flow conditions for agriculture and drinking water. However, as these supplies are utilised salinity diffuses landwards towards the tidal limit, gradually reducing the volume of freshwater immediately below the head of the estuary (MHL 1998). Water may become unsuitable for agricultural uses, human consumption and freshwater ecosystem maintenance as salinity increases (Table 1.1).

The current report investigates the occurrence of platypuses (Ornithorhynchus anatinus) in estuaries, particularly the Richmond River estuary, and investigates the tolerance of this species to saline conditions.

1.2 Distribution of Platypuses in Relation to Salinity

Platypuses normally occupy the coastal streams of eastern Australia from around Cooktown in the north to Tasmania in the south. The species is also found west and north of the Great Dividing ranges in southern Queensland, New South Wales and Victoria. However, it is most commonly found in the upper and slopes reaches of the west-flowing rivers, apart from the Murray and Murrumbidgee, where it is found, but less commonly, along the plains reaches of these rivers and their associated tributaries (Grant, 1992; Grant etal, 1999).

Over the majority of its distribution, the platypus is found in waters of low salinity (or conductivity1)- In New south Wales 43% of the rivers are reported to have salinities < 0.19 ppt (< 280 //S/cm; “Good” water quality), while 46% are between 0.19 and 0.54 ppt (280 and 800 //S/cm; “Fair” water quality), with 11% having conductivities >0.54 ppt (> 800 //S/cm; “Poor” water quality)(EPA, 1997). Most coastal rivers are classified as being in the “Good” category in terms of salinity, while the majority of upland streams are classified as at least “Fair” quality. There are several rivers however with salinities in the high range of the “Fair “ category, but in which platypuses have been reported (Grant, 1991; Grant, 1992; Grant etal, 1999). These rivers include the , a tributary of the and the , a tributary of the . The Peel River, a tributary of the and the , which is an stream of the Lachlan system both have been recorded as supporting platypuses but have salinities in the “Poor” classification. Table 1.2 shows details of these streams.

Platypuses have occasionally been found in the sea (Fleay, 1980; Connolly and Obendorf, 1998) and in estuarine habitats, but such occurrences are irregularly reported and are considered unusual (Stone, 1983; Grant, 1991; Rohweder, 1992; Hird, 1993; Menkhorst, 1995; Connolly and Obendorf, 1998). It seems unlikely that the species regularly occupies the brackish or saline waters of estuarine environments. Nothing is known of its abilities to osmoregulate under marine or brackish conditions or any need by the species to have access to fresh water to groom salt from the fur, as occurs in river (Hans Kruuk, pers. comm.). The species is known to consume a range of benthic invertebrates as food but that larvae are the most common prey items (Faragher et al, 1979; Grant, 1982). In a number of rivers along the coast of New South Wales tidal influences and/or saline intrusion into the lower reaches, results in the diversity of benthic macroinvertebrates beginning to change around the tidal limit from being numerically dominated by insect fauna to being dominated by Crustacea, including amphipods and isopods, with oligochaetes worms and gastropod molluscs also having greater

1 Electrical Conductivity (//S/cm) of water is a convenient measure of the amount of dissolved solids (salts) in the water and most sources gave such measurements rather than actual salinity (or total dissolved solids, TDS) Conductivity is affected by temperature, although most sources do not record the temperature at which measurements were made or if measurements had been corrected to a standard temperature. Conversion of Conductivity (//S/cm) to ppt (parts per thousand) in this report have been carried out by calculating: p p t= conductivity x 0.68 -4- 1 0 0 0 (ANZECC, 1992)

Appendix C - Platypus representation ( Research, 1993; Simon Williams, Australian Water Technologies, pers. comm.)- This could also have an effect on platypus distribution in the lower reaches of rivers of coastal New South Wales. It is also known that increased conductivity impairs the ability of the platypus to locate moving prey items, particularly small invertebrates, using the electrosensory mechanisms housed in its bill (Pettigrew et al, 1998 and pers. comm.).

2.0 METHODS

2.1 Platypus Distribution in Estuaries

A variety of references and data bases were searched to gain information on the occurrence of platypuses in estuarine conditions in New South Wales. These included a data base held by the author, much of which was collected during a survey of platypus distribution in New South Wales during 1987-88 (Grant, 1991), but updated since that time, the National Parks and Wildlife Service Atlas of NSW Wildlife, and the 1994-96 NSW Rivers Survey (Harris and Gerhke, 1997; Grant et al, 1999). Unfortunately funding and time constraints made it impossible to have all of the 1:100,000 map sheets covering all estuaries of New South Wales to be searched in the NPWS Atlas of NSW Wildlife and it was difficult to match known salinity data to platypus records recovered from various sources.

Rohweder (1992) carried out a Batchelor of Applied Science (Honours) project involving the study of the distribution of platypuses in the Richmond River and this document was also extensively used. Other references and personal communications are referred to in the Bibliography or in the text. In all instances records were sought in New South Wales coastal rivers downstream from either a defined (eg MHL, 1998 Figure 2 for the Richmond River) or inferred (eg tidal limit marked on a topographic map; published or unpublished salinity data) tidal limit.

2.2 Determination Salinities in Estuaries

Again a range of references, including unpublished reports (eg MHL, 1998 Figure 4 for the Richmond River) were used in an attempt to determine the salinities at, or close to where platypuses had been reported. In as many instances as possible an attempt was made to ascertain the salinity around the recorded dates of platypus observations. The number of cases where this was possible was quite low, but these are discussed in detail in the Results section. Monthly rainfall figures from Rosebank were also used for the Richmond River in an attempt to infer whether platypus observations were made during wet or dry periods (and therefore possibly high or low flows).

2.3 Investigation of Other Habitat Variables

Platypuses are known to require earth banks consolidated by the roots of vegetation as the preferred places to construct both their resting and nesting burrows (Burrell, 1927; Grant, 1995) and the occurrence of platypuses is known to be positively correlated with a number of habitat variables, including increasing length of pools and depth of water (up to 5 metres in depth), overhanging vegetation, undercut banks and the availability of habitat for benthic invertebrate food species (Rohweder, 1992; Bryant, 1993; Ellem et al, 1998; Serena etal, 1998). The data of Bishop (Appendix B; Main Report) and Bird (1997) enabled some assessment of habitat suitability in the various streams in terms of the nature of riparian vegetation present. 3.0 RESULTS

3.1 Occurrence of Platypuses in Estuaries - General Grant (1991) reported 388 positive sightings of platypuses from 191 different water bodies in New South Wales but noted only 4 from tidal or brackish areas (1%). Connolly and Obendorf (1998) mapped the distribution of platypuses in Tasmania from 630

Appendix C - Platypus 6 sightings and 69 captures, but recorded reports of the species in estuarine areas at Strahan, Penguin, Table Cape and Wynyard as “unusual”. Menkhorst (1995), commenting on the distribution of platypuses in the state of Victoria indicated that they “are absent from deep lakes and storage (>5 m deep) and from brackish estuarine waters”. In that state Fleay (1980) reported the rehabilitation of individuals from saline situations, but suggested that animals found in these situations were normally dispersing juveniles, and/or ones displaced by flood waters. In a study of the distribution of platypuses in Queensland, Stone (1983) recorded only 5 instances of platypuses being found or observed in salt or brackish habitats. He also attributed these records to the effects of floods or to juvenile dispersal, but further indicated that the common occurrence of platypuses within 10 kilometres of the tidal limit in Queensland rivers could result in their making some excursions into brackish sections of these rivers.

3.2 Occurrence of Platypuses in Estuaries - New South Wales

Platypuses have been recorded at or downstream of the tidal limits in the upper estuaries of a number of rivers in New South Wales. These data are shown in Table 3.1.

Unfortunately few salinity data were available from most of these estuaries and/or were not necessarily applicable to the times of the sightings. However these data do indicate that platypuses are found in estuaries from time to time, although the majority of these records were from the upper reaches of the estuaries2

Platypuses are regularly seen in the lower upstream of the road bridge at Bellingen, including observations in the area of tidal movement between Gravel Bar # 0 and Gravel Bar #l(Bishop, 1994; Public Works Department and MHL, 1989). However, salinity data collected by Marine Pollution Research Pty. Ltd. (1993) showed a salinity of < 0.01 ppt in this area of the river and those gathered by Public Works Department and MHL (1989) indicated that the river at that time was 0 ppt upstream of Gravel Bar #2 (~1 kilometre downstream). There have also been individual anecdotal, but unconfirmed reports made to the author of platypuses in the Bellinger River near the mouth of Hydes Creek, where salinities have been found to range between 2.5 and 8.5 ppt (Public Works Department and MHL, 1989), and downstream of Femmount, where salinities would be expected to be much higher than 18 ppt (Public Works Department and MHL, 1989).

Bishop (pers. comm.) reported a platypus immediately downstream of the tidal limit on the Hastings River during fish surveys in 1997, at a river flow of 225 ML/day. During an earlier survey in 1994 he recorded a salinity of 4 ppt at the bottom of a pool at that site (with fresh water at the surface), at a river flow of 116 ML/day. The present author also made separate observations of platypuses 1.25 and 2.25 kilometres downstream from the tidal limit in June 1999, during river flows of 979-1360 ML/day in the Hastings River. No salinity information was available for either the 1997 or 1999 observations.

One of 3 observations of platypuses in the lower noted that the sighting was in an area affected by tidal movement but that water from the river was being used to irrigate fruit trees at that time, indicating that its salinity would have been < 0.54 ppt (conductivity < 800 fiS/cm; EPA, 1997).

3.3 Occurrence of Platypuses in the Richmond Estuary Mainly due to the work of Rohweder (1992), the distribution of platypuses in the Richmond River system is probably the most well documented for any river system in New South Wales. His data, along with those from the National Parks and Wildlife Service Atlas of Wildlife of New South Wales, shows that platypuses are frequently reported around the tidal limits of the Richmond River and its tributary streams. Table 3.2 shows details of sites where platypuses have been recorded from around the tidal limits defined by MHL (1998) and downstream of these limits.

2 Estuary is defined as waters within a river system which are influenced by tidal cycles Appendix C - Platypus Platypuses were recorded as occurring up to 14 kilometres downstream of the defined tidal limit in Bungawalbin Creek, 10 kilometres downstream of the tidal limit in the Richmond River and up to 6 km downstream in the Wilsons River (upstream of Lismore). However, most records were from either close to the tidal limit or within a few kilometres downstream of that tidal limit (Table 3.2). 3.4 Matching of Salinity Data to Platypus Records in the Richmond Estuary

3.4.1 Emigrant Creek The most extreme downstream platypus record was from approximately 0.7 kilometres downstream from the tidal limit. Bishop (1999) measured a salinity of 0 ppt in Emigrant Creek to approximately 7 kilometres downstream of the tidal limit after moderate rainfall and quotes measurements of 0.09 ppt 1 km below the tidal limit in dry-weather conditions (n=l) and 0.04-0.06 ppt during wet weather (n=38), with the water being essentially “fresh” from 0.5 to 7 km downstream. Although April-September of 1992, when the work of Rohweder (1992) was carried out, were not particularly wet months, it seems unlikely that the platypus records made by him were in brackish water. This conclusion is also supported by the occurrence of typically freshwater plant species (eg water lilies, Triglochin procera) in the first 3 kilometres downstream of the tidal limit (Bishop, 1999).

3.4.2 Maguires Creek The platypus records from the lower reaches of this creek are from around the tidal limit. One of the two records was for April 1992, a fairly wet month and it seems unlikely that these platypuses would have been observed in anything but essentially fresh water.

3.4.3 Tucki Tucki Creek There is a tidal barrage on Tucki Tucki Creek at Bagotville, suggesting the platypus records in the lower part of this system would have been made from fresh water. Leakage does occur across tidal barrages (Bishop, pers. comm.), so that in the absence of published measurements it is not possible to comment accurately.

3.4.4 Bungawalbin Creek Although one record of a platypus was made 14 kilometres downstream on Bungawalbin Creek, most of the records from this creek system, including , were either close to the tidal limit or less than 5 kilometres downstream. The furthest saline penetration recorded in the Richmond estuary was 0.5 ppt at high tide during dry-weather flows on 03 November 1994 to around 60 kilometres upstream from the mouth (MHL, 1998 Figure 4). The Bungawalbin Creek-Richmond River junction occurs at approximately 55 kilometres from the river mouth. The tidal limit of Bungawalbin Creek is approximately 91 kilometres from the estuary mouth, so that the most extreme downstream platypus record would be at 77 kilometres from the mouth. As the dry weather saline influence was found to penetrate the estuary only 60 kilometres from its mouth (MHL, 1998 - Figure 4), it seems unlikely that this most extreme platypus record from Bungawalbin Creek would have been from brackish water.

3.4.5 Richmond River Although the tidal influence extends almost to Casino in the Richmond River (Keith Bishop, pers. comm.), salinity is almost certain to have been lower than 0.5 ppt at the platypus sites 9 and 10 kilometres downstream of the town (113 kilometres from the mouth).

3.4.6 Wilsons River Platypuses were recorded as occurring between Lismore (100 kilometres from mouth) and the tidal limit of the Wilsons River. As argued above for the Richmond River, platypuses

Appendix C - Platypus normally occur in fresh water upstream of Lismore (Section 3.3.5).

3.4.7 Leycester, Terania and Goolmangar Creeks

No platypus records were found for the tidal reaches of these streams, although by the argument presented above (section 3.3.5), it is unlikely that the lower reaches of these streams are brackish in nature. Their confluence with each other and that of Terania Creek with the Wilsons River are upstream of Lismore (~100 kilometres from the mouth).

Although platypuses are reported downstream of the defined tidal limits of the Richmond River and its tributaries, it appears that they are probably only occurring in waters with salinities < 0.5 ppt.

3.5 Habitat Assessment

The data of Bishop (Appendix B; Main Report) indicated that riparian vegetation upstream of the mangrove area (~45 km upstream from the mouth of the estuary in the larger streams) is fairly discontinuous along the Richmond and Wilsons Rivers but was more continuous and extensive along Bungawalbin Creek. Bishop (Appendix B; Main Report) also isolated quality ratings for riparian vegetation along various sections and arms of the Richmond River estuary from the work of Bird (1997). The upper Richmond was recorded as entirely “poor” or “moderate”. He indicated a spread of values along the Wilsons River ranging from “very poor” (~30%) to “very good” (~15%), with around 35% being classed as “good” or “moderate”. Although 20% of Bungawalbin Creek was classified as “very poor”, the rest was considered “moderate” (~33%) or “very good” (~47%) in relation to riparian vegetation.

4.0 DISCUSSION

4.1 Distribution of Platypuses and Platypus Habitat in Estuaries

It is apparent from the data that platypuses are reported from the areas of tidal influence in the upper estuarine reaches of some coastal rivers in eastern Australia, although it appears that they are only irregularly reported from reaches where the water is considered to be brackish or saline.

4.2 Determinants of Platypus Distribution in Estuaries

Although platypuses are known to occur in inland rivers which are considered to be “poor” in terms of salinity (> 0.54 ppt) or conductivity (>800 /*S/cm)(see Section 1.0 Introduction), these salinities are much less than are normally equated with brackish water in estuaries (> 1.0 ppt or 1471 piS/cm conductivity). However, it does appear that the platypus is largely restricted to the upper estuaries and the limited longitudinal salinity data available for these situations indicates that it is most frequently reported in waters where salinity is close to 0 ppt. Although the species can obviously tolerate salinities in some inland waters which are as high as 1.2 ppt (1730 //S/cm)(eg Boorowa River - see Section 1.0 Introduction and Table 1.2) and individuals of the species have been observed on occasions in more saline areas, most established populations survive and reproduce in fresh waters.

These data gathered from literature and data base searches, including information from the Richmond River estuary, indicated that platypuses are largely restricted in their distribution to the upper sections of estuaries, within a few kilometres of the limit of tidal influence. These sections of the estuary are essentially fresh water (< 0.5 ppt). The reasons for such distribution is not known, but is probably related to a number and/or combination of biotic and abiotic factors influenced by tidal cycles in the estuarine environment, rather than to the species’ tolerance to a specific level of a single factor, such as salinity. Biotic factors influencing the platypus distribution may include:

Appendix C - Platypus (a) food availability - platypuses are known to be opportunistic in their diet but predominantly consume benthic larvae of insect species in freshwater systems (Faragher et al, 1979; Grant, 1982).

(b) other resources - eg banks consolidated by vegetation suitable for burrow construction.

(c) competition for resources - eg from Fish species, such as sand whiting ( Sillago ciliata) and dusky flathead ( Platycephalus juscus), which do not normally occupy fresh water sections of rivers.

(d) predation - eg bull shark ( Carcharhinus leucus) and larger teleost species, such as Jewfish ( Argyrosomus hololepidotus), which would normally not enter the fresh water sections of rivers.

It seems likely that these factors would determined by a number or combination of tidal influences in the estuary.

4.3 Consequences of the Observed Estuarine Distribution of the Platypus

Encroachment of greater salinity further up an estuary, due to the extraction of water from the tidal pool, may not directly affect the platypus in terms of its tolerance to higher saline conditions, but may displace whatever factors are determining their distribution further upstream, thereby reducing the distribution of the species within the river system The fact that platypuses are periodically recorded in saline or brackish water in estuaries, and are found in some inland waters of “poor” quality [in terms of salinity], indicates that salinity alone may not be restricting the distribution of platypuses in estuaries.

Even under dry conditions, the penetration of salinity levels of > 0.5 ppt seems to be restricted to around 60 km from the mouth of the estuary into Bungawalbin Creek and the Richmond and Wilsons Rivers, while the tidal influence in these streams extends to around 90 (Bungawalbin Creek) to 112 km (Richmond and Wilsons River) from the mouth of the estuary. No evidence was found that platypuses in the Richmond River estuary have been recorded more than 14 km below the tidal limits in the Richmond River or its major tributaries, including the Wilsons River and Leycester, Goolmangar, Terania, Sandy, Bungawalbin, Maguires and Emigrant Creeks . Most platypus records in the estuary were around the tidal limit or less than 5 km downstream, and all were almost certainly in fresh water. In the Richmond river estuary water extraction from the tidal pools of the smaller streams would be expected to produce a greater proportional change in salinity conditions than in larger streams of the system. There is currently around 40 km of essentially fresh water between the most downstream record of platypuses in the Richmond and Wilsons Rivers but only around 4-17 km of fresh water upstream of the lowest platypus record to the tidal limit in Emigrant and Bungawalbin Creeks respectively. Water extraction from these streams would be most likely to influence platypus distribution and so ideally should not be permitted, or at least be restricted and very carefully modelled and monitored. Bungawalbin Creek is of special concern in this regard, as work shows this stream to potentially have a high habitat value for platypuses, in terms of availability of consolidated banks suitable for burrowing. The riparian vegetation consolidating these banks is much more extensive along the freshwater section of the estuarine part of this stream than along either the upper Richmond or Wilsons River estuaries (Bishop, Appendix B; Main Report; Bird, 1997). Both the Atlas of NSW Wildlife and Rohweder (1992) recorded a number of sightings of platypuses in Bungawalbin Creek downstream of the tidal limit and around the tidal limit in Sandy, Maguires and Emigrant Creeks (Table 3.2). Presumably greater volumes of water could be extracted from the tidal pools of the two main rivers of the system with less impact on the upstream environment. However, these would also need to be carefully modelled and monitoring of changes to the structure of

Appendix C - Platypus 10 flora and fauna assemblages, including the platypus, in the estuary should be carried out and adjustments made to water extraction, should adverse findings be made during the monitoring studies.

4.4 Limitations of the Current Study

Although the current study indicates that platypuses do not form resident populations more than a few kilometres downstream of the tidal limit in rivers, clearly more data on distribution and salinity need to be extracted from the literature and available data bases to investigate this conclusion. Where possible these data should be obtained from interstate, including Tasmania.

A major multi-disciplinary research project could investigate the factors which possibly determine the observed distribution of platypuses within the Richmond River estuary. However, direct observations of platypuses along canoe-based transects in the upper estuary could establish more detailed knowledge of the distribution of the platypus in the various tidal pools of the estuary. Such data would be essential as baseline information, should monitoring of platypuses be considered during or after periods of water extraction from the various tidal pools in the system.

Field work could involve the capture of platypuses. However, one of the problems of this approach, is that the method of capturing the species (unweighted nets; Grant and Carrick, 1974) often results in a large by-catch of fish species when used in the lower reaches of coastal rivers. The disturbance caused by the constant tending of the nets often results in a low capture rate of platypuses and there is an enhanced chance of drowning individuals, as they are unable to raise the nets to the surface due to the weight of trapped fish. For this reason this approach would not be preferentially recommended.

Canoe-based transect observations of platypuses, with detailed measurements of salinity along these transects (at the same time), could be used to facilitate an assessment of the use of tidal pool areas by the species in relation to patterns of salinity. Transects would need to be traversed around dawn and dusk, but could be supplemented by the use of night vision equipment. However, it must be noted that observational studies have not been widely used to date and are dependent for success on the numbers of platypuses seen per transect being fairly high (>5 animals seen /transect) during baseline studies (Grant, 1998; 1999).

5.0 CONCLUSIONS

The limited data collected from the literature and data bases searched indicate that platypuses are largely restricted in their distribution to the upper sections of estuarine waters at salinity levels which are < 0.5 ppt (735 piS/cm conductivity). It is recommended that this level of salinity be considered as the upper limit of tolerance for the species when assessing the effects of water extraction from tidal pools on salinity changes within estuaries.

However, more importantly it is concluded that the distribution of platypuses in estuaries is probably governed by the interaction of a number of possible biotic and abiotic factors, and is unlikely to be directly related only to salinity. Salinity changes, brought about by water extractions from the various tidal pools in the Richmond River estuary system, could affect these factors and it is recommended that

• any proposed extraction regimes be modelled as accurately as possible.

• an adaptive management strategy be developed, which includes strict monitoring programs and a facility for change to extraction regimes in response to either environmentally positive or negative findings of monitoring studies. • • at least initially either no extraction, or very limited extraction be made from the smaller tidal pools, including Bungawalbin and Emigrant Creeks.

Appendix C - Platypus 11

6.0 REFERENCES Australian and New Zealand Environment and Conservation Council. 1992. Australian water quality guidelines for fresh and marine waters. ANZECC.

Bird, L. 1997. The Richmond River Catchment. Stream health assessment report. An ecological health assessment of the condition of coastal streams in the Richmond River region of northern NSW. Report prepared for Department of Land and Water Conservation. November 1997.

Bishop, K.A. 1994. Coffs Harbour water supply augmentation EIS Aquatic studies - freshwater fishes - stage one - Data collection, in Raising of Karangi Dam. Environmental Impact Statement. Supplementary Documents. Report to Mitchell McCotter for NSW Public Works Department and Coffs Harbour City council.

Bishop, K.A. 1999. Revised eight-part test on 3 threatened freshwater fish species in relation to the Pacific Highway upgrading: Ballina bypass. Unpublished report to Connell Wagner Pty. Ltd. on behalf of the NSW Roads and Traffic Authority, May, 1999.

Bryant, A.G. 1993. An evaluation of the habitat characteristics of pools used by platypuses ( Ornithorhynchus anatinus) in the upper Macquarie River system. Bachelor of Applied Science (Hons) Thesis. Bathurst: Charles Sturt University.

Burrell, H. 1927. The platypus. Sydney: Angus and Robertson.

Connolly, J.H. and Obendorf, D.L. 1998. Distribution, captures and physical characteristics of the platypus ( Ornithorhynchus anatinus) in Tasmania. Australian Mammalogy, 20, 231-237. Environment Protection Authority. 1997. NSW State of the Environment. EPA, Chatswood.

Department of Land and Water Conservation. 1997. Lachlan catchment. State of the Rivers Report - 1997. DLWC, Orange.

Ellem, B.A., Bryant, A. and O’Connor. 1998. Statistical modelling of platypus {Ornithorhynchus anatinus) habitat preferences using generalised linear models. Australian Mammalogy 20,281-285.

Faragher, R.A., Grant, T.R. and Carrick, F.N. 1979. Food of the platypus, Ornithorhynchus anatinus, with notes on the food of the brown trout, Salmo trutta, in the , New South Wales. Australian Journal of Ecology 4: 171-179.

Fleay, D. 1980. Paradoxical platypus. Hobnobbing with duckbills. Jacaranda Press, Milton, Queensland.

Grant, T.R. 1982. Food of the platypus, Ornithorhynchus anatinus (Omithorhynchidae: Monotremata) from various water bodies in New South Wales. Australian Mammalogy 5: 135-136.

Grant, T.R. 1991. The biology and management of the platypus {Ornithorhynchus anatinus) in New South Wales. Species Management Report # 5. NSW National Parks and Wildlife Service, Hurstville.

Grant, T.R. 1992. Historical and current distribution of the platypus, Ornithorhynchus anatinus, in Australia, in Platypus and Echidnas (M.L. Augee, ed.). Royal Zoological Society of NSW, Sydney. Pp. 232-254.

Appendix C - Platypus 12

Grant, T.R. 1998/99. The Hastings district water supply augmentation scheme: detection of potential future water extraction impacts on aquatic biota of the lower Hastings River. Pilot monitoring: the platypus. Reports prepared K. Bishop on behalf of NSW Department of Public Works and Hastings Municipal Council. August 1998; September 1999

Grant, T.R. and Carrick, F.N. 1974. Capture and marking of the platypus, Ornithorhynchus anatinus in the wild. Australian Zoologist 18: 133-135.

Grant, T.R., Gehrke, P.C., Harris, J.H. and Hartley, S. 1999. Distribution of the platypus ('Ornithorhynchus anatinus) in New South Wales: Results of the 1994-1996 New south Wales Rivers Survey. Australian Mammalogy 21(2) in press

Harris, J.H. and Gehrke, P.C. 1997. Fish and rivers in stress. NSW Fisheries Office of Conservation and the CRC for Freshwater Ecology. Cronulla, NSW.

Hird, D. 1993. Estuarine platypus activity. Tasmanian Naturalist 144,7-8.

Manly Hydraulics Laboratory (MHL). 1998. Freshwater extraction from the Richmond River below the tidal limit. Unpublished report. Report #MHL937. DPWS Report # 98049, Manly Vale.

Marine Pollution Research Pty. Ltd. 1993. Investigation of the invertebrate benthic fauna of the Bellinger River. Unpublished report for Mitchell McCotter and Associates. June, 1993.

Menkhorst, P.W. (ed.) 1995. Mammals of Victoria. Distribution, ecology and conservation. Oxford University Press, Melbourne.

Pettigrew, J.D., Manger, P.R. and Fine, S.L.B. 1998. The sensory world of the platypus. Philosophical Transactions of the Royal Society of London B. 353, 1199-1210.

Public Works Department and Manly Hydraulics Laboratory. 1989. Bellinger River upper tidal reaches resources study. Report MLH525, April, 1989, PWD #88016.

Serena, M., Thomas, J.L., Williams, G.A. and Officer, R.C.E. 1999. Use of stream and river habitats by the platypus, Ornithorhynchus anatinus, in an urban fringe environment Australian Journal of Zoology 46,267-282.

Stone, G.C. 1983. Distribution of the platypus, Ornithorhynchus anatinus, in Queensland. Queensland National Parks and Wildlife Service, Brisbane.

Appendix C - Platypus Water Conductivity Salinity Source Use Quality fiS/cm ppt Category “Good” <280 <0.19 EPA, 1997 suitable for most irrigation purposes “Fair” 280-800 0.19-0.54 EPA, 1997 unsuitable for salt- intolerant plants eg grapes “Poor” >800 >0.54 EPA, 1997 unsuitable for a wide range of plants, including fruit, vegetable and pasture crops >350 >0.23 MHL, 1998 unsuitable for human consumption due to taste > 1500 >1.02 EPA, 1997 unsuitable for human consumption ANZECC, 1992 unsuitable for protection of freshwater aquatic ecosystems >3100 >2.11 MHL, 1997 unsuitable for stock watering

Table 1.1 Salinity guidelines for freshwaters in Australia. Electrical ConductivitypiS/cm ( ) of water is a convenient measure of the amount of dissolved solids (salts) in the water and most sources gave such measurements rather than actual salinity (or total dissolved solids, TDS) Conductivity is affected by temperature, although most sources do not record the temperature at which measurements were made or if they have been corrected to a standard temperature. Conversion of Conductivity (/*S/cm) to ppt (parts per thousand) in this report have been carried out by calculating: conductivity x 0.68 + 1000 (ANZECC, 1992)

River Conductivity Salinity Salinity Platypus Record fiS/cm ppt Record Source Source (mean; (mean; range) range) Belubula River 650; 997-233 0.44; 0.68- DLWC, 1997 Grant, 1991; Grant, 0.16 1992* Cudgegong -800 -0.54 EPA, 1997 Grant, 1991; Grant, River 1992* Boorowa River 1007; 1730- 0.68; 1.18- DLWC, 1997 Grant etal, 1999; 477 0.32 Faragher, pers. comm. Peel River 922 (median) 0.63 EPA, 1997 Grant, 1991; Grant, (median) 1992*

Table 1.2 Platypus occurrence in streams of “Poor” or very “Fair” salinity conditions in New South Wales. * associated data base held by author

Appendix C - Platypus 14

River System Waterbody Source Comment Richmond Richmond River Rohweder, 1992 See section 3.3 for Wilsons River NPWS Atlas discussion of Bungawalbin Creek available salinity Sandy Creek data Tucki Tucki Creek Maguires Creek Emigrant Creek Bellinger/Kalang Bellinger River Pers. observations See Section 3.2 for NPWS Atlas discussion of available salinity data Hastings Hastings River Pers. observations See Section 3.2 for Bishop, pers. comm. discussion of available salinity data Karuah L. Boolambayte Grant, 1991 associated data base Hawkesbury Colo River Grant, 1991 See Section 3.2 for associated data base discussion of available salinity data Bega Bega River NPWS Atlas, State Fisheries, pers. comm.

Table 3.1 Reported occurrences of platypuses in estuaries of New South Wales.

Appendix C - Platypus 15

Creek/River Location Date Source Emigrant Ck ~ just u/s and d/s 24.04-07.09.92 Rohweder - Study tidal limit Maguires Ck ~ tidal limit 18.04.92 Rohweder - Study ~ tidal limit 1992 Rohweder - Q’aire Duck Creek ~ tidal limit 1992 Rohweder - Q’aire Tucki Tucki ~ tidal limit 24.04-07.09.92 Rohweder, 1992 Creek (tidal barrage) Bungawalbin ~ 2-5km d/s tidal limit 24.04-07.09.92 Rohweder - Study Ck ~ 0.5k d/s tidal limit 31.12.85 NPWS Atlas ~ 14km d/s tidal limit 01.01.94 NPWS Atlas ~ tidal limit 1992 Rohweder - Q’aire Sandy Ck ~ tidal limit 24.OL07.09.92 Rohweder - study ~ tidal limit 1992 Rohweder - Q’aire Richmond R ~ 10km d/s tidal limit 24.04-07.09.92 Rohweder - study ~ 9km d/s tidal limit 1977 Rohweder - Q’aire Wilsons R ~ 5km d/s tidal limit 15.04.92 Rohweder - study ~ 6km d/s tidal limit 15.04.92 Rohweder - study ~ 2km d/s tidal limit 24.04- 07.09.92 Rohweder - Q’aire ~ 4km d/s tidal limit 24.04- 07.09.92 Rohweder - Q’aire ~ tidal limit 31.12.84 NPWS Atlas ~ tidal limit 1987 Rohweder - Q’aire ~ tidal limit 1991 Rohweder - Q’aire ~ tidal limit 31.12.91 NPWS Atlas ~ tidal limit 20.08.93 NPWS Atlas Leycester, only non-sightings d/s tidal 24.04-07.09.92 Rohweder - Q’aire Terania, limit Rohweder - study Goolmangar Ck

Table 3.2 Platypus records around the tidal limits in the Richmond River and its tributaries, d/s, downstream; u/s, upstream; Rohweder - Study, observations by Rohweder (1992) himself; Rohweder - Q’aire, reports received in answer to questionnaires

Appendix C - Platypus