department of defence raaf base darwin · v2 revised draft 17/09/2018 aw sr sr v3 revised draft...

Department of Defence

RAAF Base Darwin

Ecological Risk Assessment

23 October 2018

This page has been left intentionally blank

Coffey Environments Australia Pty Ltd ABN: 65 140 765 902 i

RAAF Base Darwin

Prepared for Department of Defence

Prepared by Coffey Environments Australia Pty Ltd Level 1, 436 Johnston Street Abbotsford VIC 3067 Australia t: +61 3 9290 7000 f: +61 3 9290 7499 ABN: 65 140 765 902

23 October 2018

754-MELEN199420_R09

Quality information Revision history

Revision Description Date Originator Reviewer Approver

v1 Draft 6/07/2018 GD; MB; PS AW SR

v2 Revised draft 17/09/2018 AW SR SR

v3 Revised draft 16/10/2018 AW SR SR

v3 Final 23/10/2018 AW SR SR

Distribution

Report Status No. of copies Format Distributed to Date

v1 draft 1 Word and pdf attachments

Department of Defence PFASIM 6/07/2018

v2 draft 1 Word and pdf attachments

Department of Defence PFASIM 17/09/2018

v3 draft 1 Word Department of Defence PFASIM 16/10/2018

Final 1 PDF Department of Defence PFASIM 23/10/2018

Coffey Environments Australia Pty Ltd ABN: 65 140 765 902 ii

Table of contents

1. Introduction .................................................................................................................................. 1 1.1. This Report ........................................................................................................................ 1 1.2. Background ....................................................................................................................... 1 1.3. ERA Objectives ................................................................................................................. 3 1.4. Risk Management Decisions ............................................................................................. 4 1.5. Regulatory Framework and ERA Components ................................................................. 4

1.5.1. ERA Components ................................................................................................. 5 1.6. Key assumptions and limitations of the ERA .................................................................. 10

2. Problem Identification ................................................................................................................ 12 2.1. Rationale for Undertaking the ERA ................................................................................. 12 2.2. Environmental Setting ..................................................................................................... 13

2.2.1. Surrounding Land Use ....................................................................................... 13 2.2.2. Topography and Drainage ................................................................................. 13

2.3. Sensitive Ecological Communities .................................................................................. 17 2.4. Data Evaluation ............................................................................................................... 18

2.4.1. Data Used in Risk Assessment .......................................................................... 18 2.4.2. Data Quality and Quantity .................................................................................. 19 2.4.3. Data Characteristics and Data Gaps .................................................................. 23

2.5. Conceptual Site Model .................................................................................................... 23 2.5.1. Source Areas ...................................................................................................... 25 2.5.2. Transport Mechanisms and Migration Pathways ............................................... 29 2.5.3. Exposure Pathways ............................................................................................ 30 2.5.4. Ecological Investigation Area ............................................................................. 30 2.5.5. Potential Receptors ............................................................................................ 33

3. Receptor Identification ............................................................................................................... 34 3.1. Vegetation ....................................................................................................................... 34 3.2. Aquatic Biota ................................................................................................................... 37

3.2.1. Subterranean Groundwater Fauna - Stygofauna ............................................... 39 3.3. Terrestrial Biota ............................................................................................................... 39 3.4. Receptors of Concern (ROC) .......................................................................................... 41 3.5. Potential Complete Exposure Pathways ......................................................................... 47

3.5.1. Complete Exposure Pathways ........................................................................... 47 3.5.2. Insignificant Exposure Pathways ....................................................................... 47 3.5.3. Complete Exposure Pathways Considered in Food Web Modelling ................. 48

Coffey Environments Australia Pty Ltd ABN: 65 140 765 902 iii

4. Exposure Assessment ............................................................................................................... 49 4.1. Exposure Point Concentration Selection ........................................................................ 49

4.1.1. Data Analysis ..................................................................................................... 49 4.1.2. Summary of Adopted Exposure Point Concentration ........................................ 51

4.2. Other PFAS compounds ................................................................................................. 54 4.3. Food Web Modelling ....................................................................................................... 58

4.3.1. Intake Calculations ............................................................................................. 58 4.3.2. Wildlife Exposure Parameters ............................................................................ 59

5. Toxicity Assessment .................................................................................................................. 63 5.1. Step 1 – Generic Guideline Values ................................................................................. 63

5.1.1. Surface Water .................................................................................................... 63 5.1.2. Soil ...................................................................................................................... 64 5.1.3. Sediment ............................................................................................................ 65

5.2. Step 2 – Specific Direct Toxicity Comparison – Lower Trophic Level Organisms .......... 65 5.3. Toxicity Values – Upper Trophic Level Organisms ......................................................... 68

6. Risk Characterisation ................................................................................................................ 71 6.1. Risks Associated with Direct Toxicity .............................................................................. 71

6.1.1. Terrestrial Plants and Invertebrates ................................................................... 71 6.1.2. Freshwater Plants and Invertebrates ................................................................. 75 6.1.3. Freshwater Fish .................................................................................................. 78 6.1.4. Estuarine Plants and Invertebrates .................................................................... 79 6.1.5. Estuarine Fish .................................................................................................... 79 6.1.6. Potential Contribution of Other PFAS to Direct Toxicity ..................................... 83

6.2. Risks to Higher Trophic Levels (Food Web Modelling) ................................................... 83 6.2.1. Estimated Risks to Terrestrial Receptors of Concern ........................................ 85 6.2.2. Estimated Risks to Aquatic Receptors of Concern ............................................ 97 6.2.3. Potential Contribution of Other PFAS Risks to Higher Trophic Level Organisms ..................................................................................................................... 102 6.2.4. Risk Interpretation – Food Web Modelling ....................................................... 102

7. Uncertainty and Variability Analysis ........................................................................................ 109 7.1. Limitations and Uncertainty assessment ...................................................................... 109 7.2. Parameter Sensitivity .................................................................................................... 112

8. Conclusions ............................................................................................................................. 115 9. References .............................................................................................................................. 122

Coffey Environments Australia Pty Ltd ABN: 65 140 765 902 iv

Appendices

A: Figures

Figure 1: Locality

Figure 2: Investigation Area

Figure 3: Ecological Receptors / Habitats

Figure 4: Inferred Sources of PFAS Impact

Figure 5: Surface Water Flow Plan

Figure 6: Soil Sampling Locations

Figure 7: Surface Water Sampling Locations

Figure 8: Sediment Sampling Locations

Figure 9: Terrestrial Sampling Locations

Figure 10: Aquatic Sampling Locations

Figure 11: Terrestrial and Aquatic Habitat Zones

Figure 12: Summary of PFOS Concentrations in Habitat Areas

B: Matters Protected by the Environment Protection and Biodiversity Conservation Act

C: Threatened Species Assessment (ELA, 2014)

D: Habitat Descriptions

E: Stygofauna Literature Review

F: Sampling Methods

G: Exposure Point Concentration Tables

Table 1: Soil Data

Table 2: Surface Water Data

Table 3: Sediment Data

Table 4: Terrestrial Plant Tissue Data

Table 5: Terrestrial Invertebrate Data

Table 6: Terrestrial Vertebrate Data – Mammals

Table 7: Terrestrial Vertebrate Data - Reptiles

Table 8: Terrestrial Vertebrate Data – Amphibians

Table 9: Aquatic Plant Data – Freshwater

Table 10: Aquatic Plant Data – Estuarine

Table 11: Aquatic Invertebrate Data

Table 12: Aquatic Fish Data

H: Data Statistics

I: Laboratory Reports

J: Data Validation

K: Serum to Tissue Calculations

L: Toxicity values

Coffey Environments Australia Pty Ltd ABN: 65 140 765 902 v

M: Receptor of Concern Characteristics

N: Food Web Model

O: Important Information About Your Coffey Environmental Report

Coffey Environments Australia Pty Ltd ABN: 65 140 765 902 vi

List of Acronyms 4:2 FtS 4:2 fluorotelomer sulfonate

6:2 FtS 6:2 fluorotelomer sulfonate



6:2 FtS fluorotelomer sulfonate

8:2 FtS 8:2 fluorotelomer sulfonic acid

ABS Australian Bureau of Statistics

AEC Area of Environmental Concern

AFFF Aqueous film forming foam

ASC NEPM 2013 National Environment Protection (Assessment of Site Contamination) Measure 1999 (as amended 2013)

ATSDR Agency for Toxic Substances and Disease Registry

bw Body weight

CSM Conceptual site model

CTV Critical toxicity value

DSI Detailed Site Assessment

DW Dry Weight

EEA Ecological Exposure Area

EC Effect Concentration

EFSA European Food Safety Authority

EtFOSA N-Ethyl perfluorooctane sulfonamide

EtFOSAA N-Ethyl perfluorooctane sulfonamidoacetic acid

EtFOSE N-Ethyl perfluorooctane sulfonamidoethanol

EIL Ecological Investigation Level

EPBC Environment Protection and Biodiversity Conservation Act

EPC Exposure Point Concentrations

ERA Ecological Risk Assessment

ESV Ecological Screening Value

FSANZ Food Standards Australia New Zealand

HED Human equivalent dose

HHRA Human Health Risk Assessment

IA Investigation area

IL Investigation Level

km Kilometre(s)

Coffey Environments Australia Pty Ltd ABN: 65 140 765 902 vii

LC Lethal Concentration

LOE Lines of Evidence

LOEC Lowest Observed Effect Concentration

LOAEL Lowest Observed Effect Level

LOR Laboratory limit of reporting

m Metre(s)

MeFOSA N-Methyl perfluorooctane sulfonamide

MeFOSAA N-Methyl perfluorooctane sulfonamidoacetic acid

MeFOSE N-Methyl perfluorooctane sulfonamidoethanol

MOE Margin of Exposure

mg/kg milligrams per kilogram

mg/kg milligrams per kilogram

mg/kg/day milligrams per kilogram per day

mg/L Milligrams per litre

NATA National Association of Testing Authorities

NEPC National Environment Protection Council

NEPM National Environment Protection (Assessment of Site Contamination) Measure

NHMRC National Health and Medical Research Council

NMI National Measurement Institute

NNPAS National Nutrition and Physical Activity Survey

NOEC No Observed Effect Concentration

NOAEL No Observed Adverse Effect Level

OECD Organisation for Economic Cooperation and Development

PFAS Per- and polyfluoroalkyl substance(s)

PFBA Perfluorobutanoic acid

PFBS Perfluorobutane sulfonic acid

PFBS Perfluorobutane sulfonic acid

PFC Perfluorinated compounds

PFDA Perfluorodecanoic acid

PFDoDA Perfluorododecanoic acid

PFDS Perfluorodecane sulfonic acid

PFHpA Perfluoroheptanoic acid

PFHpS Perfluoroheptane sulfonic acid

PFHxA Perfluorohexanoic acid

PFHxS Perfluorohexane sulfonic acid

Coffey Environments Australia Pty Ltd ABN: 65 140 765 902 viii

PFNA Perfluorononanoic acid

PFOA Perfluorooctanoic acid

PFOS Perfluorooctane sulfonate (alternative name Perfluorooctane sulfonic acid)

PFOSA Perfluorooctane sulfonamide

PFPeA Perfluoropentanoic acid

PFPeS Perfluoropentane sulfonic acid

PFTeDA Perfluorotetradecanoic acid

PFTrDA Perfluorotridecanoic acid

PFUnDA Perfluoroundecanoic acid

POE Point of exposure

RAAF Royal Australian Air Force

RAGS Risk Assessment Guidance for Superfund

ROC Receptor of Concern

SAQP Sampling and analysis quality plan

SSD Species Sensitivity Distribution

TDI Tolerable daily intake

TRV Toxicity reference value

UCL Upper Confidence Level

US EPA United States Environmental Protection Agency

WHO World Health Organization

WW Wet weight

μg/kg micrograms per kilogram

μg/kg/day micrograms per kilogram per day

μg/L micrograms per litre

Ecological Risk Assessment RAAF Base Darwin

Coffey 754-MELEN199420_R09 23 October 2018

1

1. Introduction 1.1. This Report As part of the Comprehensive Investigation of per- and polyfluoralkyl substances (PFAS) at RAAF Base Darwin (the Base), the Australian Department of Defence (Defence) commissioned Coffey Environments Australia Pty Ltd (Coffey) to conduct an Ecological Risk Assessment (ERA) for the Base. The purpose of the ERA being to quantitatively assess the potential for risks to the environment resulting from exposure to Base-derived PFAS in the area on and near the Base.

This ERA was prepared in accordance with the guidance provided in the National Environment Protection (Assessment of Site Contamination) Measure 1999, as amended in 2013 (NEPC 2013), including Schedule B5a Guideline on Ecological Risk Assessment, and other pertinent Schedules. Previously submitted documents provided an overview of the intended approach and methodology that was followed in the preparation of this ERA for RAAF Base Darwin.

The location of RAAF Base Darwin and the Investigation Area (IA) for this ERA are shown on Figures 1 and 2 of Appendix A.

1.2. Background Per- and poly-fluoroalkyl substances (PFAS) including per-fluorooctane sulfonate (PFOS) and per-fluorooctanoic acid (PFOA) are found in legacy Aqueous Film-forming Foam (AFFF) that were historically used to prevent or extinguish Class B fires - flammable liquid fires. AFFF is a fire-fighting foam that has been used extensively worldwide, and within Australia, from about the 1970s by both civilian and military authorities, due to its effectiveness in extinguishing liquid fuel fires. AFFF works by cooling the fire and creates a barrier between the fire’s fuel source and oxygen thereby preventing further combustion.

There are hundreds of compounds within the PFAS class, but analytical methods do not exist for all of them, and some are more prevalent than others. The active ingredient in the manufactured products (i.e. 3M Lightwater) were typically per-fluorooctane sulfonate (PFOS), per-fluorohexane sulfonate (PFHxS) or per-fluorooctanoic acid (PFOA). Newer AFFF products contain shorter chain PFAS compounds (less than six carbons) and poly-fluorinated precursor compounds (i.e. PFBS, 8:2 FTS and compounds not quantified by PFAS analysis).

The per fluorinated alkyl substances may be formed from the degradation or transformation of their precursor compounds which may be present as byproducts or intermediates from the manufacturing process and may not be able to be specifically analysed for. Therefore, on-going sources of PFOS, PFHxS and PFOA may be underestimated by analysis of these PFAS compounds alone. The full PFAS suite includes some specific known PFOS and PFOA precursor compounds.

The fully fluorinated compounds (per) do not degrade and are expected to remain in the environment for many decades. The compounds are water soluble and mobile, and will tend to migrate with water. The strength of adsorption to organic carbon in the soil of different compounds in the group varies depending on the number of carbon atoms in the compound. Long chain compounds (six or more carbons) also bioaccumulate in animals. Due to the mobility, PFAS compounds can be present in very large plumes associated with groundwater migration and surface waters. Organic rich sediments may act as ongoing or seasonal sources of PFAS contamination to surface waters through resuspension and desorption.



2

The area considered in this ERA is the Investigation Area (IA) evaluated in the RAAF Base Darwin Detailed Site Investigation (DSI) – Per- and Poly-fluoroalkyl Substances (PFAS) (Coffey 2018) (see Figure 2, Appendix A). The IA includes the RAAF Base Darwin and the surrounding area that is potentially impacted by PFAS contamination, including Rapid Creek, Ludmilla Creek, Sadgroves Creek and Reichardt Creek. The investigation area also includes Darwin International Airport (DIA) which is immediately north of the Base. The extent of the investigation area has been based on the rationale described in Table 1-1.

A separate Human Health Risk Assessment (HHRA) has been completed for the Base (Coffey, 2018). Hence this document is limited to addressing risks to flora and fauna in the terrestrial and aquatic environment that are potentially impacted by Base-derived PFAS compounds.

Table 1-1: Rationale for nomination of Investigation Area

Boundary Description Rationale

East Amy Johnson Drive Regional groundwater and surface water flow direction is towards Rapid Creek from the eastern boundary. Therefore, there is no potential pathway to direct impact from PFAS source areas on-Base to east of the Base boundary. A buffer has been provided to reflect background and upstream sampling.

North To the northern side of Rapid Creek, including the eastern part of Milner.

PFAS contamination from the Base is migrating off-Base to the north and northeast in surface water or groundwater. The boundary has been based on potential surface water and groundwater pathways, taking into account that the Ludmilla Creek and Rapid Creek catchments meet about half way along the northern boundary of the Base (McMillans Road). The Investigation Area includes the waters and sediments of Rapid Creek.

West To the coast and the southern side of Ludmilla Creek, including Bagot and Ludmilla.

PFAS contamination from the Base is migrating off-Base to the west in surface water or groundwater, which both discharge to Ludmilla Creek. The boundary has been based on potential surface water and groundwater pathways. The Investigation Area includes the waters and sediments of Ludmilla Creek.

South To the receiving waters of Reichardt Creek and Sadgroves Creek, including Winnellie and Charles Darwin National Park.

Surface water and groundwater from the southern edge of the Base migrates south and discharges to the mangroves and creeks. The Investigation Area includes the waters and sediments of Reichardt Creek and Sadgroves Creek.



3

Figure 1-1: Investigation area (IA)

1.3. ERA Objectives The objectives of the ERA were:

• To assess the potential for PFAS contaminants in soil, sediment, and surface water to pose adverse effects to ecological receptors which inhabit ecological habitats present on-Base.

• To assess the potential for reported Base-derived PFAS contaminants present in off-Base environmental media and habitats to pose adverse effects to ecological receptors that inhabit the area surrounding the Base.

• To assist in guiding the remediation and risk management measures to address PFAS contamination at the Base.

These objectives were achieved through a three-step process: • Development of a Conceptual Site Model (CSM), including a source, pathway, receptor exposure

analysis;

• Screening of maximum reported concentrations in abiotic media against adopted screening values and direct toxicity benchmarks;

• Quantitative food-web modelling where dosage of PFOS and PFOA, as the key risk driving PFAS compounds, were estimated based on consumption of contaminated dietary items (including prey and vegetation) and abiotic media for multiple receptors of concern (ROCs). The estimated dosages were then compared to adopted toxicity values to determine potential ecological risks.



4

ROC were chosen to represent feeding guilds or trophic levels because it was not feasible to predict potential adverse effects for all species that may inhabit the Investigation Area (i.e. On-Base and Off-Base areas). The ROCs chosen are described in detail in Section 3.

Species were chosen to represent the many feeding guilds that may be present at the investigation area including herbivores, invertivores, omnivores, piscivores and carnivores as well as selected to represent different taxonomic classes including birds, mammals, and reptiles. Lower trophic level organisms including fish and invertebrates are evaluated at the community level, rather than at the species level, and are assessed for direct toxicity from exposure to PFAS, and as dietary items for higher trophic levels. Identified ecological communities and habitats are presented on Figure 3, Appendix A.

1.4. Risk Management Decisions The outcomes of the ERA will inform future risk management decisions which may include the following:

• Risk communication with stakeholders.

• Further assessment of PFAS contamination to address data gaps and uncertainties of the ERA.

• Management actions to mitigate potentially complete exposure pathways to ecological receptors.

• Remediation options appraisal to address identified sources of PFAS contamination.

1.5. Regulatory Framework and ERA Components The regulatory framework that is driving this investigation stems from key pieces of environmental legislation and investigation guidelines for conducting contamination assessment on the site and surrounding area. These key pieces include the:

• National Environment Protection (Assessment of Site Contamination) Measure 1999 (NEPM); Waste Management and Pollution Control Act (the Water Act).

• PFAS National Environmental Management Plan (PFAS NEMP) (HEPA, 2018).

The NEPM provides a nationally consistent framework for identification and investigation of contaminated sites and is given effect by individual legislation and guidelines in each state and territory.

The Water Act is the primary piece of legislation that governs water resource regulation and management in the Northern Territory. Under the Water Act, beneficial uses can be declared for specific water bodies and water quality objectives are established to describe the water quality targeted to protect the relevant beneficial uses. Water Quality Objectives for the Darwin Harbour (DNR 2010) indicates that guideline values for toxicants should be sourced from ANZECC&ARMCANZ (2000) guidelines.

In January 2018, the Heads of EPAs Australia and New Zealand (HEPA) released the PFAS National Environmental Plan (PFAS NEMP). The PFAS NEMP includes updated ecological guideline values for soil, terrestrial biota guideline values, as well as freshwater and marine water guideline values for aquatic ecosystems. These updated or revised values have been incorporated into this ecological risk assessment.



5

1.5.1. ERA Components Schedule B5a of the Assessment of Site Contamination (ASC) NEPM details a step-wise progression for ecological risk assessment whereby a Preliminary ERA is developed using generic assumptions that are conservative (as used in the development of EILs) to indicate any potential for ecological risk. Where a Preliminary ERA (Level 1) indicates the potential for ecological risk under conservative assumptions, a more Definitive ERA (Level 2 and 3) may be necessary to define risks on a site-specific basis. Per the NEPM ERA guidance (NEPC 2013), an ERA (Preliminary or Definitive) consists of the following five primary components:

• Problem Identification (Section 2.0)

• Receptor Identification (Section 3.0)

• Exposure Assessment (Section 4.0)

• Toxicity Assessment (Section 5.0)

• Risk Characterisation (to support subsequent risk management and communication) (Section 6.0)

The relationships between these components are illustrated in Figure 1-2 (from NEPC 2013, Schedule B5a). The Receptor Identification, Exposure Assessment, and Toxicity Assessment components are all interrelated because the assessment of any one of these components is interdependent on the characteristics of the other two components (as per NEPC 2013). As shown in the figure, Risk Characterisation combines the information developed from the Toxicity and Exposure Assessment components.

This ERA involved conducting a preliminary ERA with progression to a Level 2 definitive ERA for key exposure pathways and receptors of concern.

The ERA was used to support risk management and risk communication activities. In accordance with the NEPM the key decision to be made at the completion of each level of preliminary and definitive ERA is to determine the need for and develop risk management controls and/or undertaken further definitive ERA works to refine the risk assessment.

Component 1: Problem Identification

This component involved the review and assessment of available site information, including data evaluation, relating to the potential PFAS impacts and determines the ecological receptors of concern that may be exposed to PFAS impacts in environmental media on-Base and off-Base. Also part of Component 1: Problem formulation was the development of the Ecological Conceptual Site Model (CSM) to determine the preliminary exposure assessments for which exposure pathways may be relevant for the identified ecological receptors of concern.



6

Figure 1-2: The five basic components of an Ecological Risk Assessment (per NEPC, 2013)

Data Evaluation

This task of the Problem Identification involved the review of available information and data from previous investigations conducted at the Base. This included the review of all analytical data related to PFAS, for identifying data gaps and limitations that are reflected in the limitations and uncertainties in the conclusions of this Preliminary ERA. The level of significance of data gaps or limitations were also noted in this section. Significance ranges from minor (i.e., limited number of samples for a certain media) to major (i.e., lack of any samples from a required media).

Ecological Conceptual Site Model Development

The ASC NEPM ERA guidance (2013) states that a conceptual site model (CSM) should be developed as part of the Problem Identification component of the ERA. The preliminary CSM is designed to diagrammatically and visually relate the exposure of receptor populations to potential source areas based upon physical site characteristics and potential exposure pathways. Important components of the preliminary conceptual model are the identification of potential sources areas, transport pathways, exposure media, exposure pathways and routes, and receptor groups. Actual or potential exposures of ecological receptors associated with the areas of PFAS use were determined by identifying the most likely pathways of contaminant release and transport. The CSM contains the following elements:

• Nature and Extent Assessment: An on-Base assessment of the nature and extent of impacts of PFAS contamination, and a determination of the likelihood of potential off-Base migration of PFAS.

• Identification of Receptors of Concern (ROCs): an evaluation of the ecology and potential receptors that may be present at the Base or in the surrounding area.

• Identification of Potential Exposure Pathways: a preliminary assessment of the potential complete and incomplete exposure pathways where ecological ROCs may be exposed to PFAS.

• Identification of Screening Levels: Guideline values and study toxicity values were identified for use in screening of direct contact exposures. Screening of PFAS was not used to eliminate areas from assessment of higher trophic level fauna, because of the nature of PFAS accumulation.



7

• Development of Food Web Model: a preliminary food web model was identified to evaluate the transfer of PFAS between trophic levels through bioaccumulation and biomagnification.

A complete exposure pathway has three components:

• a source of chemicals (stressors) that results in a release to the environment (refer Section 2.5.1);

• a pathway of chemical transport through an environmental medium (refer Section 2.5.2 and 2.5.3); and

• an exposure or contact point of the chemical and an ecological receptor (refer Section 2.5.4).

The main objective of the CSM in the ERA was to identify complete and potentially significant exposure pathways that may be present. Component 2: Receptor Identification

The Receptor Identification task determined what species may be adversely effected by Base derived PFAS, and what ecological values are to be protected within the On-Base and Off-Base areas. The local species, communities, and ecological processes of ecological value were identified based on consideration of ecological, societal, cultural, and economic factors. Selected receptors were then carried forward for evaluation in the ERA.

It is not feasible to evaluate the potential adverse effects to all the wildlife species that may occur within the On-Base and Off-Base areas. Therefore, a subset of the species (indicator species) was selected for evaluation in the ERA. These indicator species are considered to be representative of the occupied habitats and trophic levels (e.g., herbivores, carnivores) in the On-Base and Off-Base areas, as well as representative of the environmental values to be protected and of the most sensitive and potentially highly exposed species. Potential receptors are described in Section 3, and nominated receptors of concern are described in Section 3.4.

Component 3: Exposure Assessment

The Exposure Assessment section of the ERA identifies the exposure pathways that were evaluated, and the exposure point concentrations and exposure factors that were used to calculate exposures to the selected receptors. Seasonal evaluation focused on surface water since PFAS concentrations in this media can vary between wet and dry seasons. However the variation between seasons was not considered to be material (i.e. impact was present across the year), and exposure was not assessed separately for different seasons.

Exposure Units and Exposure Point Concentrations

For ecological risk assessments of large sites, it is often useful to divide the site into smaller units based on consideration of different habitats, potential levels of exposure, risk management issues, or other relevant factors. Exposure and risk estimates are then developed using the data collected for these “exposure units”.

The aquatic and terrestrial ecological exposure areas adopted are described in Section 2.5.4.

Exposure point concentrations (EPC) are the concentrations of PFAS that were used to estimate contaminant exposures to aquatic and terrestrial community receptors (e.g., fish, soil invertebrates) and the ROC species. The initial screening of potential adverse effects was conducted using the maximum PFAS concentrations detected in the exposure unit. For those exposure units with concentrations that exceed the screening guideline values, refined estimates of exposure and potential adverse effects were developed. The refined estimates used either the maximum, average or an upper estimate of the arithmetic mean concentration or the 95% UCL1, for those cases where

1 The 95% UCL is the 95 percent upper confidence limit on the arithmetic mean.



8

sufficient data are available to calculate reliable UCL concentrations. U.S. EPA’s ProUCL software Version 5.1 was used to calculate the UCLs (U.S. EPA 2016). The source of EPC data and values adopted are described in Sections 2.4 and 4.1.

Component 4: Toxicity Assessment

The Toxicity Assessment section of the ERA identified the environmental concentrations and doses of PFAS contamination that result in no harmful effects, as well as adverse effects, including chronic toxic effects. Chronic toxicity values include no-observed-adverse-effect-levels (NOAELs) and lowest-observed-adverse-effect-levels (LOAELs) and other relevant values. This ERA did not directly evaluate the potential for acute toxic effects, however this potential was considered as part of the risk characterisation where exposures exceeded chronic toxicity values. These toxicity values were used to determine the concentration of PFAS that the ecosystem can be exposed to without adverse effects or with adverse effects of a certain magnitude.

Ecological Investigation Levels (EILs) as defined in the NEPC (2013) guidelines are not available for PFAS compounds. However, generic ecological screening values from the PFAS NEMP (HEPA, 2018) have been nominated for use. These generic screening values were compared to sampling data to provide an initial indication of exposure areas that pose a potential risk and that warrant further evaluation. For areas where generic screening levels are exceeded, additional methods of estimating direct toxicity were used, such as comparison to relevant chronic toxicity values (i.e. toxicity benchmarks) for specific trophic groups as determined from searches of published ecotoxicological studies and data.

Toxicity values for PFAS based on dose (mass of chemical per unit of body weight) were identified for birds and mammals based on the results of a literature search and review of other available sources, as detailed in Section 5. This included relevant NOAELs, and LOAELs for sensitive bird and mammal species. The literature search also identified the conservative screening level toxicity values used to derive the Canadian Federal Environmental Quality Guidelines for Wildlife Diet (ECCC, 2018), these conservative screening level toxicity values are referred to as Tolerable Daily Intake (TDIs).

The toxicity values were used in the Risk Characterisation to evaluate the potential for adverse effects for the ROCs based on the dose estimates calculated from the food web modelling. Toxicity values were selected with an emphasis on chronic toxicity tests with toxicity endpoints that are relevant to the viability of wildlife populations, such as adverse effects on reproduction and offspring survival. The toxicity values selected were considered to be conservative and based on the most sensitive species, according to currently available information.

In relation to both direct toxicity and dose based toxicity, the available toxicity data were reviewed for acceptability using the methods described in Schedules B5b and B5c of the NEPM (NEPC 2013).

The literature search was conducted using:

• Organisation for Economic cooperation and Development’s (OECD) PFAS Portal and PFOS hazard assessment websites,

• U.S. EPA’s ECOTOX database (https://cfpub.epa.gov/ecotox/help.cfm?sub=about), and other relevant databases to identify and review the most current ecotoxicological information.

• Google Scholar

• Science Direct

In addition, toxicity information identified in other ERAs recently prepared for the DOD program were reviewed, including:

• Human Health and Screening Ecological Risk Assessment Fire Training Area, RAAF Tindal, Northern Territory/Kimberly (AECOM 2009),

https://cfpub.epa.gov/ecotox/help.cfm?sub=about



9

• Preliminary Ecological Risk Assessment December 2016 RAAF Base Williamtown Stage 2B Environmental Investigation (AECOM 2016); and

• Stage 2C Environmental Investigation – Preliminary Ecological Risk Assessment, Army Aviation Centre Oakey (AECOM 2016).

• DEHP-ID-183-D1 Human Health and Ecological Risk Assessment for PFAS: HMAS Albatross (EnRisks, 2017)

• Interim Human Health and Ecological Risk Assessment – RAAF Base East Sale – Per- and Poly-fluoroalkyl Substances (PFAS) Investigations (Senversa, 2017)

Component 5: Risk Characterisation

The Risk Characterisation section of the ERA combined the estimated PFAS exposures with the PFAS guideline values and toxicity values to evaluate the potential risks for the ecological receptors exposed to Base-derived PFAS contaminants. As per the definition in Schedule B5a of the NEPM, ‘Risk’ is defined as: “the probability in a certain timeframe than an adverse outcome will occur in a person, a group of people, plants, animals and/or the ecology of a specified area that is exposed to a particular dose or concentrations of a hazardous agent”.

The risk characterisation also included an analysis of the uncertainties associated with the risk estimates.

Direct Toxicity

Risks of direct toxicity to aquatic and terrestrial organisms was initially evaluated by comparing PFAS concentrations detected in environmental media to available generic screening guideline values (i.e. Tier 1 screening levels) for PFAS substances.

For terrestrial receptors this was accomplished by comparing maximum and average soil concentrations in each ecological exposure area to soil direct toxicity benchmarks for terrestrial community receptors (i.e. plants and soil invertebrates). For aquatic community receptors this was done by comparing maximum and average surface water concentrations for each ecological exposure area to direct toxicity benchmarks for aquatic organisms.

Ecological exposure areas that show potential risk were further evaluated as discussed in Section 5. This included evaluation of specific toxicity study outcomes to identify what species or types of organisms may be impacted and a discussion of the potential consequences of impacting those organisms (e.g., impacts to the food chain). Other methods of estimating direct toxicity to specific groups of organisms have been used such as the use of direct toxicity effect results. For example, the risk of direct toxicity to fish has been evaluated based on comparison of concentrations of PFOS and PFOA in water to reported sub-lethal toxic effect concentrations to fish species (i.e. No Observed Adverse Effect Concentration (NOAEC)).

Risks to Receptors of Concern (Food web modelling)

Risks to the ROCs were evaluated by comparing the estimated exposure doses from food web modelling (discussed in Section 4) to various conservative guideline values and study toxicity values (discussed in Section 5). This was used to assess the likelihood of and potential type of adverse effects to ROCs.

This was achieved by comparing the total dietary intakes for the ROCs (i.e. estimated from the food web modelling) to the toxicity values including the conservative screening toxicity value (i.e. Canadian TDI), NOAEL and LOAEL. The potential adverse effects are presented and discussed for each receptor and exposure area.



10

Uncertainty Assessment

The Risk Characterisation includes an uncertainty assessment that identifies the sources of uncertainty in the main components of the ERA and their potential impact on the risk estimates and conclusions of the ERA.

1.6. Key assumptions and limitations of the ERA The ERA is based on the following key assumptions and limitations:

• The assessment of ecological risk is predominately based on data obtained in 2017 and 2018. The estimation of historical exposures is not possible as the historic concentrations of PFAS in various media is largely unknown.

• PFOS and PFOA have been selected as the key Contaminants of Potential Concern (COPC) as PFOS is the most prevalent compound detected in environmental media within the investigation area and guideline values are available for PFOS and PFOA. PFAS compounds other than PFOS and PFOA have been detected in various media, these have generally represented a minor portion of the total PFAS concentration. Ecotoxicological information available for these other PFAS compounds is limited, this represents a key limitation in evaluating the potential ecological risks associated with these other PFAS compounds.

• The investigation area for the ERA includes RAAF Base Darwin and the surrounding area that is potentially impacted by Base-derived PFAS including Rapid Creek, Ludmilla Creek, Sadgroves Creek and Reichardt Creek. This includes both freshwater and estuarine portions of the Creeks. The receiving marine environment in Darwin Harbour or Beagle Gulf have not been directly assessed.

• The environmental sampling of abiotic and biotic media was targeted and focussed on source areas and the nearest environmental receptors. The environmental sampling likely included the ‘worst case’ concentrations rather than the ‘typical’ concentrations.

• We have conducted food web modelling to evaluate potential risks associated with upper trophic level organisms. It was not feasible to predict risks to all species, as such indicator species were selected as receptors of concern (ROC). The indicator species were chosen to represent different feeding guilds, trophic levels and key sensitive species considered most likely to present in the investigation area.

• Data on PFAS in shallow soils <0.5 m bgl have been included in the ERA as this upper horizon is considered most relevant to exposures to terrestrial ecological receptors.

• PFAS in soils > 0.5 m bgl are considered more likely to pose an ecological risk associated with leaching to groundwater and subsequent discharge to surface waters. Groundwater concentrations have not been directly used in the ERA as the ecological risks associated with groundwater contamination have been assessed based on exposures in which groundwater discharges. Concentrations directly measured in surface water bodies have been used to assess this pathway.

• The ERA has not utilised data on PFAS in edible plant material collected from home and community gardens as part of the HHRA, as these are not considered to be key dietary sources for herbivores within the investigation area.

• Statistical analysis has been used to determine the exposure point concentration (i.e. based on 95% Upper Confidence Limit (UCL) and/or average concentrations). Non-detects have been excluded from the statistical analysis, which is considered to be a conservative approach and likely to provide a higher estimate of exposure concentrations, which is likely over-estimate risk. Similarly the maximum concentration has been conservatively adopted for small data sets (< 9 samples).



11

• No avian tissues were sampled from the investigation, as such, for the food web modelling reptile tissue data has been used as a surrogate where bird tissue was a dietary component. This is on the assumption that reptiles and small birds have a similar diet, hence the accumulated tissue concentrations would be similar. Therefore intake through consumption of birds has a low reliability in the absolute value, however, as reptile tissue was reported with higher concentrations than other animals sampled, and birds are more likely to have a larger feeding range than reptiles, the estimate is considered to be conservative and likely to over-estimate exposure.

• The wildlife exposure parameters used for the food web modelling, including dietary ingestion rates, diet composition, and body weights, have been based on best available information from the literature and professional judgement. It is acknowledged that there is uncertainty around the wildlife exposure parameters adopted in the absence of site-specific evaluations and that published values adopted are generally based on organisms from temperate rather than tropic climates.

• Conservative study toxicity values have been adopted based on the lowest no-adverse effects levels reported for the most sensitive species identified from the currently available toxicity data. It has been assumed that these conservative toxicity values will overestimate the risk to receptors of concern. However, this is an area of uncertainty, therefore the risk to receptors of concern may be under estimated for sensitive species.

• In the absence of a guideline toxicity value or study toxicity values for reptiles the toxicity value for avian species has been adopted. Therefore, the assessment of risk to reptiles has low reliability.

• It has been assumed that the indicator species considered in the food web modelling only consume PFAS contaminated dietary items. This is considered to be a conservative approach given that PFAS concentrations in dietary sources will vary across the ecological exposure area and testing has likely been biased towards more highly contaminated areas, particularly for terrestrial habitats.

• We have assumed that receptors forage all their prey and dietary items from across a single defined ecological exposure area (i.e. 100% of diet from only one area). This may over-estimate the risk potential for receptors that have a larger foraging range, particular where the foraging range includes unimpacted areas.



12

2. Problem Identification The Problem Identification presented in this section provides a synopsis of the need for this ERA at the Base, the environmental setting, sensitive ecological communities that may occur within the investigation area, a summary and evaluation of the data that are available and used in the ERA, the Ecological Investigation Levels (EILs) used to conduct the ERA, and the Conceptual Site Model (CSM) that identifies the complete and incomplete pathways by which ecological receptors may be exposed to Base-derived PFAS contaminants.

2.1. Rationale for Undertaking the ERA The rationale for undertaking the ERA has been determined from the preliminary and detailed site investigation (DSI) works that have been conducted at the site during 2017 and 2018 (Coffey, 2018a).

The site history review identified that Aqueous Film Forming Foam (AFFF) containing PFAS has been used at RAAF Base Darwin for fire fighter training and emergency fire-fighting response. Confirmed and suspected PFAS contamination sources have been identified at RAAF Base Darwin and confirmed source areas are shown in Figure 4, Appendix A.

As such during 2017 and 2018 site investigation works have been undertaken to investigate the potential nature and extent of PFAS contamination at the Base and surrounding areas. The investigation is part of a broader national program being undertaken by Defence to manage this legacy issue (see http://www.defence.gov.au/Environment/PFAS/).

The site investigation works conducted at the site during 2017 and 2018 identified contaminated soil, groundwater and surface water within both on-Base and off-Base areas. As a preliminary evaluation of potential ecological risks, associated with Base derived PFAS, the concentrations of PFOS and PFOA reported in soil, groundwater and surface water were compared to Tier 1 ecological guideline values that have been approved for use (Section 5.0). In summary, this identified that PFAS, primarily as indicated by PFOS, exceeded guideline values and warranted further evaluation for the following receptors:

• Aquatic species in on-Base waterways.

• Freshwater aquatic species in Rapid Creek (predominantly off-Base).

• Estuarine aquatic species in Ludmilla Creek and Rapid Creek (off-Base).

• Terrestrial ecosystems on-Base.

The preliminary evaluation of ecological risks indicated that PFOS presented a negligible risk to marine ecosystems in Ludmilla, Sadgroves and Reichardt creeks. Riparian ecosystems related to Rapid Creek and Ludmilla Creek have the potential for exposure to PFAS, however a relevant investigation level was not identified for screening in the DSI.

Biota sampling, including freshwater fin-fish and crustaceans, was undertaken in the off-Base creeks as part of the DSI works. This identified the presence of elevated concentrations of PFAS with the biota sampling and hence the potential for ecological risks associated with bioaccumulation.

Based on the results of the comparisons to Tier 1 Screening Levels, Coffey recommended that an ERA be conducted, as presented in this report, to assess the potential for adverse ecological effects associated with Base derived PFAS and to inform subsequent risk management decisions.

Subsequently, further environmental investigations into the nature and extent of PFAS contamination within the IA were conducted as part of the Detailed Site Investigation (DSI) (Coffey, 2018a) for which to inform this ERA. Biota sampling was also conducted to provide targeted data to support the Human Health Risk Assessment (HHRA) and ERA based on potential pathways identified in the DSI.

An overview of the PFOS concentrations reported in soil, surface water and sediment through the DSI and subsequent investigations are presented in Figures 6, 7 and 8 in Appendix A.

http://www.defence.gov.au/Environment/PFAS/



13

2.2. Environmental Setting RAAF Base Darwin is 1,278 hectares (ha) in area and is bound by both urban and industrial areas (Appendix A, Figure 1). The Base is bound by Stuart Highway, Bagot Road, McMillians Road and Amy Johnson Avenue. RAAF Base Darwin is located on Commonwealth freehold title and comprises Sections 5060 and 3437 (Hundred of Bagot).

The Base broadly comprises the following features and structures:

• Aircraft movement areas;

• Explosive ordnance storage and loading facilities;

• Command, control and communication facilities;

• Aircraft maintenance facilities;

• Aviation Fuel Farm and support facilities;

• Domestic facilities;

• Navigational aids;

• Married and single accommodation with messes; and

• Open space comprising sporting ovals and vegetation.

Air movement areas (runways and taxiways) are utilised by both civilian air operations and Defence. Darwin International Airport is classified as a Jointly Used Airport under the Airports Act 1996 (Cwlth).

2.2.1. Surrounding Land Use Land use in the area immediately surrounding the Base is summarized in Table 2-1.

Table 2-1: Surrounding land use summary

Area Land Use

North Darwin International Airport, Marrara Sporting Complex, Residential housing and future development.

East Grassland/shrub land, some light industry and the Berrimah Power Station.

South Berrimah and Winnellie industrial estates; and Charles Darwin National Park (south of Winnellie)

West Ludmilla primary school, Bagot community housing, RAAF Golf Course, residential housing.

2.2.2. Topography and Drainage

RAAF Base Darwin is relatively flat and low-lying, compared to surrounding land. It is gently sloping at a height of 10 to 33 m Australian height datum (AHD) and slopes down along the north boundary and to the southwestern corner.

Surface water on the base flows to drainage systems that consist of large open unlined drains, municipal drains and underground piping. The off-Base water bodies that receive water from the Base are Rapid Creek to the east and north, Sadgroves and Reichardt creeks to the south, and Ludmilla Creek to the west.

The topography and drainage at the Base and across the investigation area is shown on Figure 2-1 and described further below.



14

Base Open Surface Drainage Network

The Base is drained by large open unlined drains, municipal drains and underground piping that discharge surface waters into Rapid Creek on the north and east sides of the Base via Marrara Swamp; Sadgroves Creek on the south; and Ludmilla Creek on the west (refer Figure 5, Appendix A). A small central portion of the Base on the southern boundary discharges to Reichardt Creek.

Several water retention ponds have been constructed on RAAF Base Darwin to prevent flooding of both the runway and external roads. The golf-course lake near the western boundary also provides minor retention of stormwater. There are two drainage lines that run parallel to the major runway and feed runoff into the smaller drainage network of RAAF Base Darwin.

Rapid Creek

The Rapid Creek catchment encompasses the eastern and northern portion of the Base, and is the predominant water catchment extending across approximately 70% of the Base. Rapid Creek is the closest surface water body to the Base, it is approximately 9.8 km in length, drains a catchment of approximately 30km2, and flows in a general north-westerly direction.

Marrara Swamp located at the eastern end of the Base form the headwaters of Rapid Creek. The swamp splits into two main channels (either side of the Explosive Ordnance Storage Facility (NT0239)), which meet just before a flood control weir to the immediate north of Darwin International Airport. During the wet season (November – April) the creek’s flow is driven primarily by rainfall run-off whilst dry season (May-October) flow is largely dependent on groundwater discharge (Dostine, P.L, 2014). It is understood that water quality within Rapid Creek at the start of the wet season is poor due to the influx of pollutants (including bacteria, pesticides, nutrients, hydrocarbons and sediments) from the catchment washed into to creek with the first rains, though as the wet season continues water quality improves (AHU, 2006).

The RAAF Base drainage network within the Rapid Creek catchment feeds surface water runoff towards Rapid Creek via a series of mostly unsealed open drains. In the south-eastern portion of the Base the area near the redundant fire training area (NT0241) has a slightly higher elevation than the surrounding land, with an unlined drainage path flowing adjacent to the Base moving water from northwest to southeast to a slight depression where water is held in the wet season. Further northward surface water flows are guided towards a prominent open spoon drain, which also captures runoff from the eastern end of the runway, and flows in a northerly direction towards the swamp headwaters of Rapid Creek. Further northwest (across most of the extent of the runway), surface water flows are northward towards DIA where they enter drains that flow northward to Rapid Creek. A recently upgraded drain network on DIA captures runoff into a major drain on the western side of the main terminal area, which flows northward beneath Charles Eaton Drive and discharges to Rapid Creek.

In the north-western portion of the Base in the former fire training area (NT1030) and current fire training area (NT0243) predominant surface water flows are to the north towards Osgood Drive where flows are directed eastward along open spoon drains, and north east across DIA via minor drains that interconnect with more prominent drains that direct flows to the north east beneath Charles Eaton Drive and into Rapid Creek. This drainage network traverses modified woodland habitat.



15

Ludmilla Creek

The Ludmilla Creek catchment incorporates the western portion of the RAAF Base, from the northwest in the area of the skid pan, the area beyond the western end of the runway, and in the south west portion of the Base, the area around Hangar 31 (NT1031) and the former RAAF Fire Station. Drainage occurs mainly as sheet flow that collects in small, unlined drains before discharging into several drainage systems under Bagot Road.

In the northern portion of the Base, to the north of the skid pan, surface water flows are predominantly westward towards perimeter drainage that flows to the south before being diverted westward via drains that flow beneath Bagot Road, the most northerly of which is located northwest of the skid pan. Several open drains intersect the golf course along the western boundary of the Base, which divert to subsurface drains beneath Bagot Rd. These drains capture flow from the western end of the secondary runway (which runs north-south) and area near Hangar 31 via recently upgraded (2014-2015) drain network that divert flow to the north west into an on-Base retention basin and toward the drains that flow westward beneath Bagot Road. A dam within the golf course is recharged from surface water runoff from the surrounding area, but is not fed by the major drain upgrade (that collects water from the area including the surrounds of Hangar 31).

In the southern portion of the Base within the Ludmilla Creek catchment, in the area near the former Mechanical Equipment Operational Maintenance Squadron (MEOMS) building (NT0215), surface drainage flows are also westerly towards subsurface drains beneath Bagot Road.

To the west of Bagot Road, most of the drains resurface as open and generally unlined drains, except for the lined major drain that runs beside Fitzer Road to the north, and subsurface drainage that flows beneath the Bagot community. Further westward, drains flow beneath Dick Ward Drive, before discharging into Ludmilla Creek, which discharges into Darwin Harbour and the Beagle Gulf.

Sadgroves Creek / Reichardt Creek

The Sadgroves Creek and Reichardt Creek catchment encompass a smaller southern portion of the Base. This includes an area to the south of the OLAs (NT0242), and incorporates at least part of the area where the former Darwin Airport operated, and a sports field near the entrance to the Base.

Drainage occurs mainly as sheet flow (in the area south of the OLAs) that collects in small, unlined drains that direct flow southwards towards Stuart Highway, where water is further directed to sub-surface drains beneath the highway. To the south of Stuart Highway within the Winnellie area, drainage is mostly sub-surface (except for a large open drain to the east of the area which passes under the railway line). To the south of Tiger Brennan Drive, several subsurface drains resurface and discharge water into Sadgroves Creek to the west, and Reichardt Creek to the east.



16

Figure 2-1: Topography and drainage at the investigation area



17

2.3. Sensitive Ecological Communities To determine whether matters of national environmental significance or other matters protected by the Environment Protection and Biodiversity Conservation Act 1999 (EPBC Act) are likely to occur within the IA, a search was conducted using the Protected Matters Search Tool (http://www.environment.gov.au/epbc/pmst/index.html). The search was conducted July 17, 2017 and used a search radius of 20 km, approximately centred on RAAF Base Darwin. The results are provided in Appendix B and briefly summarised below.

For matters of national environmental significance, the search results indicated that within 20 km of the Base there are:

• No World Heritage Properties, National Heritage Places, Wetlands of International Importance, Commonwealth Marine Areas, or Listed Threatened Ecological Communities.

• 40 Listed Threatened Species (13 birds, 11 mammals, 7 reptiles, 7 sharks, and 2 plants).

• 69 Listed Migratory Species (6 migratory marine bird species, 23 migratory marine species, 6 migratory terrestrial species 34 migratory wetland species).

Regarding other matters protected by the EPBC Act, the following results were provided:

• there were no Critical Habitats, Terrestrial Commonwealth Reserves, or Marine Commonwealth Reserves.

• 28 possible occurrences of Commonwealth Land.

• 7 Commonwealth Heritage Places.

• 111 Listed Marine Species (55 species of birds, 1 mammal species, 27 species of reptiles, and 28 species of fish).

• 12 whales and other cetaceans.

The species identified above have been considered to determine which of these species occur or are likely to occur within the investigation area and would be appropriate to be assessed as potential receptors, as discussed in Section 3.

In addition to the species identified by the Protected Matters Search Tool, information was also sourced from plant and animal species listed under the Northern Territory Territory Parks and Wildlife Conservation Act of 2001 (https://nt.gov.au/environment).

The Preliminary Draft Master Plan for Darwin International Airport indicates that seven threated species have been recorded at DIA (DIA 2017). These include: Darwin cycad (Cycas armstrongii), northern quoll (Dasyurus hallucatus), floodplain monitor (Varanus panoptes), Mitchell’s water monitor (Varanus mitchelli), black-footed tree-rat (Mesembriomys gouldii), pale field rat (Rattus tunneyi), and curlew sandpiper (Calidris ferruginea). The Darwin cycad and black-footed tree-rat were recorded during a 2013-2014 field survey.

ELA’s Ecological Assessment (ELA 2014) for F-35 aircraft operations, identified threatened species of in the RAAF Base Darwin F-35 study area that could be affected by the program and the occurrence of those species is further described in Appendix C.

http://www.environment.gov.au/epbc/pmst/index.html



18

2.4. Data Evaluation 2.4.1. Data Used in Risk Assessment

Data Sources

This assessment of the potential ecological risk has included data and information from previous reports including:

• Coffey 2018 – Detailed Site Investigation – Per- and Poly-fluoroalkyl Substances (PFAS), RAAF Base Darwin (Coffey, February 2018 Project Ref: 754-MELEN199421-R05).

• Coffey 2018 – Human Health Risk Assessment, RAAF Base Darwin (Coffey, June 2018, Project Ref: 754-MELEN199421-R07).

• GHD 2016 - Waste Classification and PFAS Investigation Report, RAAF Base Darwin, NT (GHD, June 2016 Project Ref: 4322496).

• GHD 2012 - Report for DEHP Stage 3 Fire Training Upgrade – RAAF Darwin Options for Improvement and Concept Design, RAAF Base Darwin, NT (GHD, August 2012 Project Ref: 21/21200).

• URS 2011 - NT2067 Fuel Farm 1 Bioremediation Stage 1 Site Assessment Report - DRAFT, RAAF Darwin, NT (URS, 7 February 2011 Project Ref: 42213949/R001/1).

• GHD 2010a - Remediation Part C & Part F, Sampling Analysis and Quality Plan, RAAF Base Darwin, NT (GHD 2010).

• GHD 2010b - Crash Site Investigation, RAAF Base Darwin, NT (GHD September 2010 Project ref: 43/21737/30419).

• ERM 2007 - Stage 2 Environmental Investigations NT/K-, RAAF Base Darwin, NT (ERM June 2007 Project ref: 0042239 RP4 Stage 2 Darwin).

Summary tables have been developed for all the collected data to show sample identification, and the following information for each location: number of samples analysed, number of detections for PFAS, range of detected concentrations for each PFAS, range of detection limits for the non-detected results, and relevant laboratory qualifiers. These tables are included as Tables 2-5 through 2-12. Full data sets are presented in Appendix G.



19

2.4.2. Data Quality and Quantity

Data Quality

The analytical data for abiotic (soil, sediment, surface water and pore water) and biotic media used in this preliminary ERA have been obtained primarily from Coffey (2018), GHD (2016, 2012, 2010a, b), URS (2011), and ERM (2007) reports. Coffey implemented a comprehensive quality assurance/quality control (QA/QC) program as part of our field soil, surface sediment, surface water and biota sampling procedures, based on relevant Australian Standards, EPA Guidelines and industry practice.

The implemented QA/QC program included the following:

• The use of appropriately qualified/trained environmental scientists to conduct the assessment.

• The use of standardised field records to document the findings of the assessment.

• Appropriate preservation of samples during transport from the field to the laboratory.

• The use of chain of custody documentation to ensure the traceability of sample transport and handling.

• The use of laboratories accredited by the National Association of Testing Authorities Australia (NATA) for the analysis of samples.

• The collection and analysis of field quality control samples.

• Review of internal analysis of laboratory quality control samples.

• The use of appropriate laboratory reporting limits.

• Compliance with sample holding times.

• Comparison of field and analytical data to check for the occurrence of apparently unusual or anomalous results.

The approach is generally based on guidance from the following sources:

• NEPC (2013) National Environment Protection (Assessment of Site Contamination) Amendment Measure 2013 (No. 1).

• HEPA (2018) PFAS National Environmental Management Plan, January 2018.

• AS4482.1 (2005) Guide to the sampling and investigation of potentially contaminated soil, Part 1: Non Volatile and Semi-volatile Substances.

• AS4482.2 (1999) Guide to the sampling and investigation of potentially contaminated soil, Part 2: Volatile Substances.

• AS/NZ 5667.1 (1998) Water Quality Sampling – Guidance on the design of sampling programs, sampling techniques and the preservation and handling of samples.

• ANZECC (2000) Australian and New Zealand guidelines for fresh and marine water quality.

• ANZECC (2000a) Australian guidelines for water quality monitoring and reporting.

• WA DER (2017) Interim Guideline on the Assessment and Management of Perfluoroalkyl and Polyfluoroalkyl Substances (PFAS).

The results of the data validation undertaken as part of the DSI report (Coffey, 2018a) and HHRA report (Coffey, 2018b) are contained in those reports. Full lab reports and review of data quality specific to this ERA are presented in Appendix I and J respectively.



20

Non-conformances with data quality acceptance objectives were mainly associated with the relative difference in the results from field and secondary duplicate (primary and secondary laboratory check samples, respectively) analysis. More non-conformances were recorded for soil and sediment samples, than water samples indicating the greater potential for variability of concentrations within the media sampled, which was primarily considered to be due to the inherent heterogeneity of these media. Variability was also noted between aquatic biota samples, but is consider to reflect the variability of the sample concentrations, rather than issues with laboratory analysis. This highlights the need for sufficient samples to be collected and analysed to produce statistical values for potential exposure concentrations, where elevated concentrations are indicated. As a number of elevated relative percentage difference (RPD) results were recorded in the field and secondary duplicates, more than laboratory duplicates, this indicated it was unlikely to be a laboratory precision issue.

Overall, the data are acceptable for the purposes of undertaking a preliminary ecological risk assessment.

Data Quantity

The investigations listed in Section 2.4.1 collected samples of soil, sediment, surface water and biota (aquatic and terrestrial) from on- and off-Base. A summary of the quantity of data utilised in the current ERA is provided in Table 2-3. It is important to note that sample numbers presented in Table 2-3 include QC duplicate samples for biota assessment.

It is noted that groundwater data have not been considered in the current ERA as surface water data are considered to be most relevant to exposures to ecological receptors. No further discussion of groundwater data has been included in the current ERA. Overall the quantity of data was considered appropriate for undertaking the current ERA.



21

Table 2-3: Summary of data quantity and source

Media or Biota Number of Sampling Locations (Samples)

Source Reference

Data Utilised in ERA

Soil 762 (824) Coffey 2018 Coffey 2017 GHD 2016 GHD 2012 URS 2011 GHD 2010

Only shallow soil samples (<0.5 m bgl) have been used in the data analysis.

Sediment – Freshwater 26 (29) Coffey 2018 Coffey 2017

All sediment samples have been used in the data analysis

Sediment – Estuarine 14 (15) Coffey 2017 All sediment samples have been used in the data analysis

Surface Water – Drains Freshwater

51 (143) Coffey 2018 Coffey 2017

All surface water samples have been included in the data analysis

Surface Water – Streams Freshwater

21 (115) Coffey 2018 Coffey 2017

All surface water samples have been included in the data analysis

Surface Water -Streams Estuarine

12 (67) Coffey 2018 Coffey 2017

All surface water samples have been including in the data analysis

Vegetation 16 (52) Coffey 2017 All terrestrial, freshwater and estuarine vegetation samples considered to be a potential dietary source for herbivories have been included in the data analysis. We have not included data from edible plant material collected from home and community gardens as part of the HHRA

Terrestrial Invertebrates 7 (30) Coffey 2017 Composite samples of terrestrial invertebrates collected have been considered

Mammals 5 (16) Coffey 2018 All mammal samples collected were included in the data analysis. This include whole organisms and tissue samples (i.e. flesh, organs). Serum data collected from small mammals considered to be ecological receptors of concern was used in the ERA. Serum data from domestic animals was not utilised.

Reptiles and Amphibians

7 (63) Coffey 2018 All reptile and amphibian samples collected were included in the data analysis. This included whole organisms, composite samples and in some instances targeted tissue samples.

Aquatic Invertebrates - Freshwater

6 (77) Coffey 2017 All tissue samples collected were included in the data analysis. This included whole samples and tissue samples. The whole samples were flesh / tissue samples with shells removed.

Aquatic Invertebrates - Estuarine

7 (69) Coffey 2017 All tissue samples collected were included in the data analysis. This included whole samples and tissue samples. The whole samples were flesh / tissue samples with shells removed.

Fish – Freshwater 9 (64) Coffey 2018

Coffey 2017

All tissue samples collected were included in the data analysis. This included whole samples and tissue samples.

Fish – Estuarine 7 (83) Coffey 2018

Coffey 2017

All tissue samples collected were included in the data analysis. This included whole samples and tissue samples.



22

Data Exclusions

Table 2-4 summarises any data that was excluded and the rationale for the exclusion from the current ERA.

Table 2-4: Summary of data excluded and rationale for exclusion from ERA

Excluded Data Rationale

Groundwater Groundwater concentrations have not been directly used in the ERA. This is as the ecological risks associated with groundwater contamination have been assessed based on exposures in the surface water bodies in which groundwater discharges.

Soil > 0.5 m bgl PFAS concentrations above the screening criteria are predominately in soil less than 0.5 m bgl and this upper soil horizon is considered most relevant to exposures to terrestrial ecological receptors. PFAS in soils deeper than 0.5 m bgl is considered more likely to pose an ecological risk associated with leaching to groundwater and subsequent discharge to surface waters. This exposure pathway has been assessed by way of evaluation of PFAS in surface waters, and associated sediments.

Edible plant material from home and community gardens

We have not directly used some vegetation samples that were collected for the purposes of the HHRA, specifically where these vegetation samples were considered to be edible plant material from home and community gardens (e.g. figs, limes, raspberries, spinach, basil, banana, sweet potato). These such vegetation samples were specifically collected for the purpose of the HHRA and are not considered to be key dietary sources for herbivores within the EEA.



23

2.4.3. Data Characteristics and Data Gaps

Sampling of fish and other aquatic biota was undertaken primarily for data collection to support the HHRA. Terrestrial and aquatic invertebrates, reptiles and amphibians, and small mammals were sampled for the ERA. Some data limitations were identified as part of this process and have led to adoption of potentially over-conservative exposure assumptions. These exposure assumptions may lead to over-estimation of potential risks. Table 2-5 indicates the limitations and the potential impact on the estimation of ecological risk.

Table 2-5: Summary of data limitations

Data Characteristic Potential Impact of Data Gap or Limitation on ERA

Exposure Assumption Used in ERA to Address Data Gap or Limitation

Small data sets for some specific areas or media including:

• Aquatic invertebrates from Rapid Creek Freshwater.

• Terrestrial invertebrates and Plants in Zone C and D EEAs.

• Freshwater and estuarine/marine aquatic plants from all EEAs.

• Reptiles from all terrestrial EEAs.

• Amphibians from Zone D EEA.

• Mammals from Zone B, C and D EEAs.

contained less than nine samples. This is not an ideal sample size for conducting meaningful statistical analysis.

The small sample size for some media / biota has resulted in the use of maximum reported values as conservative EPCs, which may over-estimate the exposure of ecological receptors.

The use of conservative estimates (i.e., maximum or average detected concentration) where the data set is too small to conduct statistical analysis.

No avian tissues were sampled from the investigation area as part of the data collection.

There is potential that the assessment of exposures using some surrogate measure (i.e. reptile tissue data) may over-estimate the potential ecological intakes, as reptile tissue concentrations were the highest of biota sampled.

Reptile tissue data has been used to assess potential exposures because of the similarity in diet of small reptiles and small birds.

2.5. Conceptual Site Model As part of the ERA the Conceptual Site Model (CSM) was developed to identify and illustrate the exposure pathways (both complete and incomplete) by which ecological receptors may be exposed to PFAS compounds on-Base and off-Base (Figure 2-2). The CSM was developed in accordance with Schedule B2 in National Environment Protection (Assessment of Site Contamination) Measure 1999, as amended in 2013 (NEPC 2013) and considered source areas of contamination, the pathways by which contaminants may migrate through environmental media on-Base and off-Base, the exposure pathways to receptors, and the types of ecological receptors that may be exposed both on-Base and off-Base.



24

Potential Source Source Details Likely Primary Release Mechanism Potentially Affected Watershed Potentially Affected Media Potential Exposure Pathway

Aquatic Terrestrial

Fish

\mph

ibia

In

vert

ebra

tes

Plan

ts

Bird

s

Mam

mal

s

Rept

iles

Inve

rteb

rate

s

Plan

ts

Bird

s

Mam

mal

s

Rept

iles

Current Fire Training Area

PFAS compounds have entered the soil and surface water in the area of the site as a result of direct spray of AFFF during training activities and migrated with stormwater flow.

Direct Exposure (Dermal contact and respiration)

X X X - - - - - - - - Former Fire Training Ground 1

PFAS compounds have been released at the site during AFFF training and testing activities.

Former Fire Training Ground 2

PFAS compounds have been released at the site during AFFF training and testing activities Bioconcentration –

Uptake through Food Chain

X - - X X X - - X X X Airservices Fire Station PFAS containing AFFF may have been discharged to the environment by leaks, spills, or direct application of AFFF or wash water to the soil or drains in the vicinity.

Former Aviation Rescue and Fire Fighting Station

PFAS compounds may have been discharged to the environment in this area as a result of direct discharge of AFFF to surface soil in the training area or leaks and spills from storage of AFFF.

Incidental Ingestion (while feeding/foraging)

X - - X X X - - - - -

Former Fuel Farm 4 Leaks or discharge from the AFFF fire suppression system at the Fuel Farm. Ingestion (as drinking

water) - - - X X X - - X X X

Former Fuel Farm 5 Leaks or discharge from the AFFF fire suppression system at the Fuel Farm. Direct Exposure

(Dermal Contact) - X X - - - - - - - - Former Fuel Farm 6 Leaks or discharge from the AFFF fire suppression system at the Fuel

Farm. Stockpiled AFFF Contaminated Soils

Soils contained in lines stockpiles following remediation of a release of AFFF at Hanger 31 in March 2008. Incidental Ingestion

(while feeding/foraging)

X - - X X X - - - - - Former Darwin International Airport (DIA) Terminal

A fuel spill incident that involved the discharge of AFFF over the spill site and down a stormwater drain where fuel had migrated.

Sports Oval Reported AFFF application was not found to have contaminated this area.

Bioaccumulation – Uptake through Food Chain

X X - X X X - - X X X Former Fuel Farm 1 Leaks or spills from storage of AFFF as well as accumulation in

shallow soils and migration to surface waters and sediments.

MEOMS / GEMS Workshop 542

Release to the environment as a result of contaminated wash-down water, or from the emptying of fire truck tanks that contained liquids containing PFAS was not found to have contaminated this area.

Former RAAF Fire Station

PFAS compounds may have been discharged to the environment in this area as a result of direct discharge of AFFF to surface soil in the training area or leaks and spills from storage of AFFF. Direct Exposure

(Foliar, Root or Dermal Contact)

- - - - - - X X - - - Former Defence Burn Pits and Landfills

Incident response or testing along the runway, application of AFFF to fuel spills, use of foam for recreational purposes or equipment washing, and discharge of handheld foam extinguishers.

Fred Smith Sports Field Reported AFFF application was not found to have contaminated this area. Incidental Ingestion

(while feeding/foraging)

- - - - - - - - X X X Hangar 31 – AFFF The results of several incidents associated with the automated foam

suppression system. Air North Aircraft Crash Site

Reported AFFF application was not found to have contaminated this area.

Bioaccumulation – Uptake through Food Chain

- - - - - - X - X X X Former Runway / Skid Pan

Reported AFFF application was not found to have contaminated this area.

Other diffuse or unidentified sources (i.e., incident response or testing along the runway, application of AFFF to fuel spills, etc.)

Figure 2-2: Conceptual Site Model for PFAS exposure to receptors of concern at RAAF Darwin

Seepage

Seepage

Infil

tratio

n

Infiltration

Infil

tratio

n

Seepage

Direct Application

Direct Application

Adso

rptio

n D

esor

ptio

n

Rapid Creek

Surface Water

Sediment

Surface Soils Surface

Soil

Direct Spray

Leaks/spills

Ludmilla Creek

Leaching

Direct Spray

Leaks/spills

Leaks/spills

Surface Soil

Sadgroves and

Reichardt creeks

Surface Soil

Ground water

Ground water

Ground water



25

2.5.1. Source Areas

Source areas have been identified across the investigation area related to use, storage or discharge of PFAS-based contaminants. Identified source areas are as follows:

The locations of these source areas are indicated on Figure 2-3 and Figure 4 of Appendix A.

Figure 2-3: Potential PFAS source areas

Current Fire Training Area

The current Fire Training Ground, which is also referred to as the Hot Fire Training Ground, is on RAAF Base Darwin on the boundary of the Darwin International Airport. The fire training area is operated by Airservices Australia (ASA) and commenced use in 2002. PFAS compounds have entered the soil and surface water in the area of the Base as a result of direct spray of AFFF during training activities and migrated with stormwater flow and groundwater flow. Based on the reported concentrations and anticipated migration pathway, the potentially relevant ecological receptors from contamination from the current Fire Training Ground are terrestrial flora and fauna at the Base and aquatic flora and fauna in Rapid Creek.



26


The former Fire Training Ground is at the southeastern end of the main runway and the former facility covered an area of approximately 18,300 m2. The site was used for fire training prior to 1981 (based on anecdotal evidence) until the late 1990s, though a definitive date for cessation of training is not known. Fire training involved use of water, protein based foams and 3M LightwaterTM (AFFF)2 (based on the period) and dry chemical powder. PFAS compounds have been released at the site during AFFF training and testing activities over a period of approximately 20 years. Contaminants may have adsorbed to surface soils or migrated with water and sediments in stormwater. Based on the reported concentrations and anticipated migration pathways, the potentially relevant ecological receptors include terrestrial flora and fauna at the site and aquatic flora and fauna in Rapid Creek.


This area was composed of an earthen Fire Training Ground, which was used for training by ARFF and Defence from 1970 to 1990. Protein based foams were used prior to introduction of synthetic AFFF products in 1978. By 1982 protein foams had been phased out and 3M LightwaterTM was consistently used. PFAS compounds have been released to the surface soils at the site during AFFF training. Contaminants may have adsorbed to surface soils or migrated with water and sediments in stormwater. Based on the reported concentrations and anticipated migration pathway, the potentially relevant ecological receptors include terrestrial flora and fauna at the site and aquatic flora and fauna in Rapid Creek.

Airservices Fire Station

The Airservices Fire Station is on the northern side of the main DIA runway. Activities at the station include storage and decanting of AFFF, fire fighting vehicle parking, maintenance and washdown, equipment testing and training. Airservices ceased the use of 3M LighwaterTM in 2003 when AFFF was transitioned to Ansulite3. Airservices has since ceased using Ansulite (2010) but Ansulite is still stored onsite and made available for emergency response. PFAS containing AFFF may have been discharged to the environment by leaks, spills, or direct application of AFFF or washwater to the soil or drains in the vicinity of the Airservices Fire Station. Contaminants may have adsorbed to surface soils or migrated with water and sediments in stormwater. The Airservices fire station is not on RAAF Base Darwin and has not been operated by the Department of Defence, and therefore it was not characterised as part of the RAAF Base Darwin PFAS investigation. However, the potential ecological receptors of aquatic flora and fauna in Rapid Creek have been assessed through direct sampling at the receiving water body.

2 Chemical composition of 3M’s Light Water (FC-203CF) AFFF includes 69.0 – 71.0% Water; 20% diethylene glycol butyl ether (butyl carbitol); 1.0 – 5.0% amphoteric fluoralkylamide derivative; 1.0 – 5.0% alkyl sulfate salts; 0.5 – 1.5% perfluoralkyl sulfonate salts (PFOS); 0.5 – 1.5 triethanolamine; and 0.05% tolytriazole (corrosion inhibitor) (Moody and Field, 2002). 3 Chemical composition of Ansulite incudes >80% Proprietary mixture consisting of hydrocarbon surfactants, fluorosurfactants, inorganic salts, and water; 19% diethylene glycol monobutyl ether; 0.2% tertiary butyl alcohol, 0.3% hexylene glycol.



27

Former RAAF Fire Station

The RAAF fire station was located at building 558 on Bukatilla Road and included a fire alarm system monitoring, vehicle parking, equipment storage and AFFF storage. The fire station was operated from prior to 1974 until approximately 1990, when all fire services responsibilities transitioned to Aviation Rescue and Fire Fighting (ARFF) operated by the Civil Aviation Authority at the time, and later Airservices Australia. PFAS compounds may have been discharged to the environment in this area as a result of direct discharge of AFFF to surface soil in the training area directly northeast of building 558 or leaks or spill from storage of AFFF. Contaminants may have adsorbed to surface soils or migrated with water and sediments in stormwater. Based on the reported concentrations and anticipated migration pathway, the potentially relevant ecological receptors from the former RAAF fire station include terrestrial flora and fauna at the site and aquatic flora and fauna in Ludmilla Creek.

Former Aviation Rescue and Fire Fighting Station

A former Aviation Rescue and Fire Fighting (ARFF) Station, operated by Airservices post-1995, and previously by Civil Aviation Authority and the Commonwealth Department of Civil Aviation, was located on the south side of the runway, near the current OLAs. The station was built in 1958 (Brown & Root 2000) and operated until approximately 1999 when Airservices moved to a new fire station on Darwin International Airport. PFAS contaminants may have been discharged to the environment in this area as a result of direct discharge of AFFF to surface soil or leaks or spills from storage of AFFF. Contaminants may have accumulated in shallow soils or migrated in surface water or sediments (likely to be in a northerly direction overland). Based on the potential for PFAS to have been discharged in the area and anticipated migration pathway, the potentially relevant receptors from contamination from the former ARFF fire station include terrestrial flora and fauna at the site and aquatic flora and fauna in Rapid Creek.

Former Fuel Farm 1

Former Fuel Farm 1 was located south of Hangar 31 and had a capacity of 212 kL (HLA 2004). The facility was constructed in the 1940s and decommissioned prior to 2011 (URS 2011). The site infrastructure included fuel tanks, associated infrastructure, and unlined containment bund, fire pump, water tank and a moveable 1 kL tank of AFFF. PFAS contaminants has been inferred to have been discharged to the environment in this area as a result of leaks or spills from storage of AFFF. Contaminants may have accumulated in shallow soils or migrated in surface water or sediments (likely to be in a northerly direction overland). Based on the potential for PFAS to have been discharged in the area and anticipated migration pathways, the potentially relevant receptors from contamination from the former ARFF fire station include terrestrial flora and fauna at the site and aquatic flora and fauna in Ludmilla or Sadgroves Creek.

Former Fuel Farm 4

Former Fuel Farm 4 was located on the southern side of the main runway and was constructed in the 1960s. The farm had a 1,200 kL capacity in above ground tanks within an earthen bund, and supplied Fuel Farms 5 and 6 (ERM, 2007). PFAS contamination has been inferred to have been released to the environment as a result of leaks or discharge from the AFFF fire suppression system at the Fuel Farm. Where released, contaminants may have adsorbed to shallow soils, migrated with water or sediment in stormwater or infiltrated to groundwater. Based on the reported concentrations and anticipated migration pathway, the potentially relevant ecological receptors from contamination at Fuel Farm 4 include terrestrial flora and fauna at the site and aquatic flora and fauna in Rapid Creek.



28

Former Fuel Farm 5

Former Fuel Farm 5 was located on the southern side of the main runway, at the south-eastern end. The farm had a capacity of 306 kL and serviced the fighter replenishment area (FRA) (ERM 2007). The farm was decommissioned in approximately 2012. PFAS contamination has been inferred to have been released to the environment as a result of leaks or discharge from the AFFF fire suppression system at the Fuel Farm. Where released, contaminants may have adsorbed to shallow soils, migrated with water or sediment in stormwater or infiltrated to groundwater. Based on the reported concentrations and anticipated migration pathway, the potentially relevant ecological receptors from contamination at Fuel Farm 5 include terrestrial flora and fauna at the site and aquatic flora and fauna in Rapid Creek.

Former Fuel Farm 6

Former Fuel Farm 6 was located on the southern side of the main runway and was constructed in the 1960s, and was supplied with fuel by Fuel Farm 4, approximately 150 m east (ERM, 2007). PFAS contamination has been inferred to have been released to the environment as a result of leaks or discharge from the AFFF fire suppression system at the Fuel Farm. Where released, contaminants may have adsorbed to shallow soils, migrated with water or sediment in stormwater or infiltrated to groundwater. Based on the reported concentrations and anticipated migration pathway, the potentially relevant ecological receptors from contamination at Fuel Farm 6 include terrestrial flora and fauna at the site and aquatic flora and fauna in Rapid Creek.

Stockpiled AFFF Contaminated Soils

A release of AFFF at Hanger 31 in March 2008 of up to 5,000 L resulted in impacts to the adjacent earthen/grassed surface area. The top 100 – 150 mm of soil in the worst affected areas of the release were stripped off and stockpiled in an area to the immediate south-west of Hanger 31. Stockpiled soils were removed in June 2015 and relocated to a designated landfill but in time, contaminants may have leached from the stockpile and may adsorb to surface soils or migrate with water and sediments in stormwater. Based on the reported concentrations and anticipated migration pathway, the potentially relevant ecological receptors from contamination of stockpiled soil include terrestrial flora and fauna at the site and aquatic flora and fauna in Sadgroves Creek.

Hangar 31 – AFFF

Hangar 31 is a large steel framed and clad aircraft hangar with some adjoining smaller brick and mortar buildings used as offices. The hangar contains an automated foam system to provide after-hours suppression of any fuel fires. The installation of the foam suppression system is understood to have occurred in or prior to 1985 during which time synthetic AFFF products were in use. PFAS compounds have been released at the site as a result of several incidents associated with the automated foam suppression system over the past 30 years and contaminants may have adsorbed to surface soils or migrated with water and sediment in stormwater. Based on the reported concentrations and anticipated migration pathway, the potentially relevant receptors from contamination from the Hangar 31 area include terrestrial flora and fauna at the site and aquatic flora and fauna in Ludmilla Creek.



29

Former Defence Burn Pits and Landfills

Other diffuse or unidentified sources may also be present, such as incident response or testing along the runway, application of AFFF to fuel spills, use of foam for recreational purposes or equipment washing, and discharge of handheld foam extinguishers. These areas have not been specifically assessed. The cumulative effect of these potential sources is being evaluated through investigation of the main site drains (surface water and sediment) and site-wide groundwater investigation.

Secondary sources of PFAS on-Base include, but are not limited to:

• PFAS adhered/adsorbed to soils in the vadose zone, migrating to groundwater or surface water with seasonal fluctuations in groundwater levels;

• Elevated PFAS concentrations dissolved in the underlying groundwater beneath source areas;

• PFAS in sediments along Rapid, Ludmilla, Reichardt, and Sadgroves Creeks, drainage lines and surface soils across the IA mobilising during the wet season; and

• Movement of contaminated soils during historic construction activities and potential generation of PFAS-laden dust.

2.5.2. lTransport Mechanisms and Migration Pathways The potential transport mechanisms and migration pathways for PFAS contamination within the IA are the following:

• Surface water runoff to site drains, open pits, and Rapid Creek, Ludmilla Creek, Reichardt Creek and Sadgroves Creek;

• Migration in surface water in site drains and Rapid Creek, Ludmilla Creek, Reichardt Creek and Sadgroves Creek as shown in Figure 5, Appendix A;

• Infiltration of surface water to subsurface soil and groundwater;

• Leaching from soils and sediments into groundwater and surface water;

• Lateral migration in groundwater;

• Discharge of groundwater into surface water and associated sediments;

• Uptake of PFAS in sediment, soil, groundwater and surface water by biota;

• Uptake of PFAS in higher trophic level organisms by ingestion of biota; and

• Release of PFAS to the environment from the decomposition of deceased biota that have accumulated PFAS in their tissues.



30

2.5.3. Exposure Pathways Based on the current understanding of the impacted media and the ecological receptors that may occur on-Base and in the affected off-Base areas the following exposure pathways are potentially complete and are evaluated in this ERA:

Surface Soil

• Direct contact to soil invertebrates.

• Uptake and accumulation by terrestrial plants.

• Incidental ingestion by wildlife while foraging.

• Ingestion by wildlife of animals and plants that have accumulated PFAS substances from impacted surface soil.

Sediments (in Rapid Creek, Ludmilla Creek, Reichardt Creek and Sadgroves Creek)

• Direct contact with sediment invertebrates.

• Incidental ingestion by wildlife while foraging.

• Ingestion by wildlife of aquatic animals and plants that have accumulated PFAS substances from impacted sediments.

Surface Water (including Rapid Creek, Ludmilla Creek, Reichardt Creek and Sadgroves Creek)

• Direct contact with fish, aquatic invertebrates, amphibians and aquatic plants.

• Water ingestion by wildlife.

• Ingestion by wildlife of animals and plants that have accumulated PFAS substances from impacted surface water.

2.5.4. Ecological Investigation Area

A preliminary ERA is an iterative process that can be refined as new data is collected, and that enables areas of concern to be focused on as the ERA is refined. The preliminary ERA is used to inform future sampling, risk assessment and management decision but is not intended to provide overall final conclusions with respect to potential risks to ecological receptors at the Base and surrounding areas.

This preliminary ERA includes the delineation of the Investigation Area into different ecological exposure areas (EEAs) and risks to ecological receptors are assessed for each EEA. For this preliminary ERA, entire IA was evaluated with respect to terrestrial, freshwater, and estuarine/marine habitats but then was also sub-divided into smaller EEAs with respect to the evaluation of the terrestrial food web and the aquatic food web.



31

Terrestrial Ecological Exposure Areas

The Base was divided into four quadrants with the surface water drains acting as the driver for division. Figure 2-4 illustrates the division of the investigation area into the four quadrants represented by Zones A, B, C and D (also presented on Figure 11, Appendix A). Figure 3 of Appendix A shows the vegetation communities. Further description of vegetation and habitat is provided in Appendix D along with photos of the areas.

• Zone A

The Zone A EEA represents the northeast quadrant and encompasses the upper portion of Rapid Creek including the northern and southern headwater tributaries to Rapid Creek. Terrestrial habitat is expansive across this area and the woodland has no public access and relatively minor disturbance. Fauna surveys in 2012 found one amphibian, 36 birds, six mammals and eight reptile species residing in this area (ELA, 2012). The terrestrial habitat in this EEA is part of a contiguous bushland area spanning the length of Rapid Creek. There are no known source areas within Zone A, but surface waters are impacted from sources within Zone D via groundwater migration and surface runoff.

• Zone B

The Zone B EEA represents the northwest quadrant and encompasses middle and lower portions of Rapid Creek and eucalyptus woodland on Base (Figure 3 and Figure 11, Appendix A). The terrestrial habitat in this EEA is part of a contiguous bushland area spanning the full length of Rapid Creek, with small breaks at road corridors. Fauna are relatively free to move within this corridor. The on-Base woodland was in reasonably good condition with moderate to dense tree cover and scattered shrubs linking Rapid Creek to the golf course. Known source areas within or potentially impacting this EEA include: former Fire Training Area 2 (NT0242), current Fire Training Ground (NT0243) and potentially the current Airservices Fire Station.

• Zone C

The Zone C EEA represents the southwest quadrant and encompasses the main operating part of the Base and the area draining to Ludmilla Creek. Habitat for native fauna is limited on Base but may include planted trees, shrubs and grasses around buildings and in the golf course. The trees would provide good bird roosting areas and refuge from the operating areas and open-grass fairways. The terrestrial habitats of the middle portion of Ludmilla Creek were a mix between mangroves and Paperbark, Pandanus, and Eucalypt forests. Known source area within or potentially impacting this EEA include Hangar 31 (NT1031); former Fuel Farm 1 (NT0201); and former RAAF Fire Station.

• Zone D

The Zone D EEA represents the southeast quadrant and encompasses facilities in the southern part of the Base, and the main airstrip. The area drains to Rapid Creek to the north via site drains and to the south to Reichardt Creek. The terrestrial habitat in this EEA consisted of native woodland patches and weeds, with a variety of tree size classes (except very large trees). Fauna are able to move between patches by crossing unsealed tracks and grassland. The closest accessible waterway is the headwaters of Rapid Creek. Known source areas within or potentially impacting this EEA include: former Fire Training Area 1 (NT0241); former Fuel Farm 5 (NT0206); former Fuel Farm 4 (NT0205); former Fuel Farm 6 (NT0207); and the AFFF contaminated soil stockpiles (NT1002).



32

Aquatic Ecological Exposure Areas

The aquatic EEAs included the major creek systems within the investigation area including:

• Freshwater reach of Rapid Creek

This section includes the Marrara Swamp and Rapid Creek through Pandanus swamps and Monsoon Rainforest. The headways and upper reaches are ephemeral and would not provide sustained aquatic habitat. The creek becomes perennial from approximately the weir and is fed by groundwater flowing from either side of the creek, including from beneath RAAF Base Darwin. Trees overhang the creek and the benthic substrate is clay and large aggregate in upstream sections, and sands, silt and detritus in sections downstream of the weir. Banks were frequently undercut with exposed roots. PFAS contamination from the Base enters the creek via surface water runoff, groundwater discharge in creek beds and via drainage from deep earthen drains that intercept groundwater.

• Estuarine portion of Rapid Creek

This EEA includes the permanent estuarine reaches of Rapid Creek, nominally downstream of Trower Road. The creek typically comprised steep mud banks and dense mangroves. Gastropods were evident on trunks and roots and the area provides good habitat for fish and foraging species. PFAS contamination from the fresh water reaches of Rapid Creek enter this section and are diluted by tidal flushing.

• Ludmilla Creek

Ludmilla Creek is a permanent estuarine water way that receives fresh stormwater and groundwater discharge from the surrounding suburbs (including RAAF Base Darwin). The creek is surrounded by mangrove forest and typically has gentle mud banks in upper reaches and sand and gravel flats towards the mouth. Deep channels are present in some areas and rock substrate is exposed in areas of constricted flow in upper reaches. Gastropods were observed on tree trunks and roots and the creek and surrounds generally provides good habitat for fish, foraging animals and birds, although the southern back has been modified near the mouth for the boat ramp and facilities. PFAS contamination from Zone C on-Base ultimately discharges to Ludmilla Creek via stormwater discharge and groundwater seepage into drains and is diluted by tidal flushing.

• Reichardt and Sadgroves creeks

Reichardt and Sadgroves creeks are tidal arms of Darwin Harbour which also receive fresh water run-off and stormwater from Winnellie and the southern portion of RAAF Base Darwin. The freshwater sections do not represent viable habitat. The habitat in the estuarine reaches fluctuated with the tides and included soft marine muds and flooded mangroves. Occasional discharge of PFAS contaminated stormwater enters these creeks from piped stormwater drains. Minor groundwater seepage may also contribute to PFAS impact in upper reaches of the creeks.

A combined estuarine environment was also considered, which represents aquatic fauna that moves within Beagle Gulf and Darwin Harbour or piscivores and omnivores that collect food from the wider area.



33

Figure 2-2: Terrestrial and aquatic habitat zones

2.5.5. Potential Receptors A variety of animals and plants found in aquatic and terrestrial habitats may be exposed to PFAS compounds both on-Base and off-Base. The species and ecosystems that are associated with the IA are identified in the Receptor Identification section of this ERA (Section 3.0).



34

3. Receptor Identification Receptor Identification determines what species or families should be evaluated in the ERA based on potential exposures and what ecological values are to be protected at the Site. The local species, communities, and ecological processes of ecological value were identified based on consideration of ecological, societal, cultural, and economic factors. Selected receptors were carried forward for evaluation in the ERA. Below is a brief summary of the local ecology and species that occur in the IA and where PFAS concentrations have been assessed. Figure 3-1 presents the food web for the region.

A habitat survey was conducted prior to sampling to further identify the local ecosystems, habitats, and species that may be exposed to site-derived PFAS contamination and refine the specific species targeted for analysis. The habitat description and selected photos are presented in Appendix D.

Soil and water sampling and analysis has been reported in the DSI report (Coffey 2018a) and subsequent events were reported in the HHRA report (Coffey 2018b). Fish and aquatic invertebrate sampling and analysis was reported in the HHRA report (Coffey 2018b). The sampling methods for Terrestrial biota (plants, invertebrates and vertebrates) are presented in Appendix F of this report, and laboratory reports and data quality review are presented in Appendix I and J respectively.

3.1. Vegetation The Base and surrounding area fall within the Australian tropical savanna region which covers a substantial portion of northern Australia. The savanna is characterized by a wet season that begins in October and lasts through March and a dry season from about April through September. Based on data from Darwin airport, most of the rainfall is between January and March (ELA 2014). Wildfires are a significant feature of the dry season and have a considerable influence on the landscape and the types of plants and animals that occupy the area. The terrestrial environment at Darwin falls within the H12 vegetation classification unit developed by Fox et al. (2001 in ELA 2014) that is characterized by knotted box (Eucalyptus persistens) and/or silver-leaved ironbark (E. shirley) open-woodland with sparse ground cover comprised of tussock grasses (ELA 2014). Forests, swamps, and Eucalypt woodlands occur in the Marrara Swamp and in the riparian zone of Rapid Creek (SKM 2009b in ELA 2014).

As part of the sampling conducted in November 2017 (wet season), many types of vegetation were observed and some were sampled on and near the Darwin Base for concentrations of PFAS. Observations made in field are summarised in Table 3-1 and described in Appendix D.



35

Table 3-1: Summary of vegetation species observed on and near the Darwin Base in November and December 2017

Terrestrial Vegetation Common Name (Species Name)

Terrestrial Vegetation Common Name (Species Name)

Aquatic Vegetation Common Name (Species Name)

Breynia (Breynia cernua) Carallia (Carallia brachiata) Beach Hibiscus (Hibiscus tiliaceus)

Broad-leaved Paperbark (Melaleuca viridiflora)

Northern Black Wattle (Acacia auriculiformis)

Frogsmouth (Philydrum lanuginosum)

Knotweed (Pericaria sp.) Black Speargrass (Heteropogon sp.)

Water Lily (Nymphoides sp.)

Terminalia (Terminalia erythrocarpa) Mixed grasses Freshwater Mangrove (Barringtonia acutangula)

Syzgium (Syzgium angophoroides) Gamba Grass (Andropogon gayanus)

Stilt Rooted Mangrove (Rhizophora stylosa)

Weeping Fig (Ficus benjamina) Darwin Stringybark (Eucalyptus tetrodonta)

Saltmarsh (sp 1)

Cluster Fig (Ficus racemosa) Rosella (Hibiscus sp.) Saltmarsh (sp 2)

Pandanus (Pandanus spiralis) Poinciana (Delonix regina) Grey Mangrove (Avicennia marina)

Turkey Bush1 (Calytrix exstipulata) Milkwood 1 (Alstonia actinophylla)

Padina (Padina sp.)

Bloodwood 1 (Corymbia bleeseri) Coffee bush 1 (Leucaene leucocoephala)

Mangrove Apple (Sonneratia alba)

Cycad 1 (Cycas armstrongii) Medicine Paperbark 1 (Melaleuca cajuputi)

Maranthes 1 (Maranthes corymbosa) Soap tree 1 (Alphitonia excels)

Green plum (Buchanania obovate) Ghost gum 1 (Corymbia bella)

Ironwood 1 (Erythrophleum chlorostachys)

Sand palm 1 (Livistona humilis)

1 – Vegetation was not sampled for concentrations of PFAS

.



36

Figure 3-1: Food web diagram for RAAF Base Darwin and surrounds



37

3.2. Aquatic Biota The primary aquatic environments within the IA are Rapid Creek on the north and east sides of the Base, Sadgroves Creek and Reichardt Creek are on the south side of the Base, and Ludmilla Creek is on the west side.

Rapid Creek drains a catchment of approximately 30km2 and is approximately 9.8 km long. It flows northwesterly and discharges into Darwin Harbour. The upper portion of the creek is freshwater and the lower portion, approximately 2.4 kilometres, is tidal and experiences extreme tidal inundation. Rapid Creek has designated beneficial uses of “aquatic ecosystem protection” and “recreation and aesthetics” under the Northern Territory Water Act (NRETA, 2010). The mangroves in the region are considered ecologically significant and provide habitat for birds, and a variety of estuarine and marine fish, crustaceans and other organisms (Fortune 2015). Sadgroves Creek and Reichardt Creek flow south into Charles Darwin National Park. Ludmilla Creek is an extensive salt and freshwater wetland system that includes areas of coastline, rainforest, mangrove and woodlands. The creek flows to the west and discharges to Beagle Gulf.

Rapid, Ludmilla, Sadgroves and Reichardt creeks support a variety of fish and aquatic invertebrates (Tables 3-2 to 3-5). Both saltwater crocodiles (Crocodylus porosus) and freshwater crocodiles (C. johnstoni) inhabit Rapid Creek.

Table 3-2: Summary of fish and invertebrate species identified and collected in the freshwater reaches of Rapid Creek during sampling activities by Coffey conducted in March, April, and November 2017

Fish Common Name (Species Name)


Invertebrates Common Name (Species Name)

Spangled Perch (aka spangled Grunter) (Leiopotherapon unicolor)

Western Rainbowfish (Melanotaenia australis)

Red Claw Crayfish (Cherax quadricarinatus)

Hyrtl’s Catfish (Neosilurus hyrtlii) Purplespotted Gudgeon (Mogurnda mogurnda)

Chequered Rainbowfish (Melanotaenia splendida inornata)

Tarpon / Oxeye Herring (Megalops cyprinoides)

Blackbanded Rainbowfish (Melanotaienia nigrans)

Table 3-3: Summary of fish and invertebrate species identified and collected in the estuarine reach of Rapid Creek during sampling activities by Coffey conducted in November 2017 and March 2018



Barramundi (Lates calcarifer) Mud Crab (Scylla serrata)

Goldspotted Rockcod (Epinephelus coioides) Long Bum (Telescopium telescopium)

Pikey Bream (Acanthopagrus pacificus) Mangrove crabs (Perisesarma spp.)

Forktail Catfish (Neoarius sp.) Orange Mud Crab (Scylla olivacea)

Banded Archerfish (Toxotes jaculatrix) Periwinkle (Nerita balteata)

Garfish (Hemiramphidae) Hermit crabs (Pagroidea)

Tarpon (Megalops cyprinoides)



38

Table 3-4: Summary of fish and invertebrate species identified and collected in Ludmilla Creek during sampling activities by Coffey conducted in March, April, and November 2017 and the University of Queensland (Vijayasarathy et al. 2017)




Barramundi (Lates calcarifer) Mangrove Jack (Lutjanus argentimaculatus)

Mangrove Swimming Crab (Thalamita crenata)

Sea Mullet (Mugil cephalus) Longfin Silver Biddy (Pentaprion longimanus)

Giant Mangrove Whelk(Terebralia palustris)

Brown Sweetlips (Plectorhinchus gibbosus)

Pikey Bream (Acanthopagrus pacificus)

Mud Crab (Scylla serrata)

Flathead (Platycephalidae) Garfish (Hemiramphidae) Periwinkle (Nerita balteata)

Scat (Scatophagus sp.) Glassfish (Ambassis sp.) Oysters and Cockles

Giant Tiger Prawn(Penaeus monodon)

Long Bum (Telescopium telescopium)

Sand Crab (Portunus pelagicus)

Hermit Crab(Paguroidea (super family))

Table 3-5: Summary of fish and invertebrate species identified and collected in Darwin Harbour (including Sadgroves Creek) during sampling activities by Coffey conducted in November 2017 and March 2018




Tarpon (Megalops cyprinoides) Banded Archerfish (Toxotes jaculatrix)

Giant Mangrove Whelk (Terebralia palustris)

Golden Snapper (Lutjanus johnii) Flathead (Platycephalidae) Tiger Prawn (Penaeus monodon)

Giant Queenfish (Scomberoides commersonnianus)

Blue Salmon (Eleutheronema tetradactylum)

Sand Crab (Portunus pelagicus)

Sicklefin Lemon Shark (Negaprion acutidens)

King Salmon (Polydactylus macrochir)

Periwinkle (Nerita balteata)

Bull Shark (Carcharhinus leucas) Goldspotted Rockcod (Epinephelus coioides)

Mud Crab (Scylla serrata)

Barramundi (Lates calcarifer) Barred Javelin (Pomadasys kaakan)

Orange Mud Crab (Scylla olivacea)

Trevally (Caranx sp.) Long Bum (Telescopium telescopium)



39

3.2.1. Subterranean Groundwater Fauna - Stygofauna As part of the receptor identified phase of this ERA Coffey conducted a desktop review of the potential for subterranean groundwater fauna, in the form of stygofauna, to be present at the site the findings from this review are summarised below.

Stygofauna are invertebrates that live in subterranean groundwater systems. There are hundreds of identified species of stygofauna across the world and in recent years the semi-arid aquifers in Australia have been identified as significant and unique ecosystems. Determination of the location and diversity of stygofauna communities is challenging as investigations must be conducted through discrete boreholes, or springs and the sampling methods may influence the results of assessments. Desktop studies can be applied to assess the suitability of aquifers as stygofauna habitat, however the actual presence of communities relies on connection or historic introduction of species into that aquifer system. Stygofauna typically require fresh recharge aquifers, which contain oxygenated water with organic matter leached from shallow soils as a food source (Humphreys 2000; Hancock and Boulton 2008). Aquifers with connected voids, such as karst systems or fractured rock networks have been identified as high diversity habitats. Humphreys (2000) identified that in some karstic aquifers, stygofauna may be present at depths exceeding 100m, due to the high vertical exchange of groundwater drawing oxygen and organic matter deep into the aquifer. Stygofauna require relatively stable water conditions (salinity, pH, temperature and water presence) and fluctuations in water levels that would result in dewatering of porous aquifer zones, such as mine dewatering or over extraction are a significant threat to stygofauna communities. Chemical contamination of aquifers is also a threat (Commonwealth of Australia 2009).

A selection of studies conducted across Australia are summarised in Appendix E. In the Northern Territory, stygofauna has been confirmed through published reports in saturated sands beneath Magela Creek and in springs from the Karns Dolomite at Pungalina Station. Anecdotal information and unpublished reports also indicated the presence of stygofauna in Cutta Cutta caves, springs in Katherine River. Studies in other parts of Australia have identified many species of stygofauna in alluvium and karst aquifers in the Pilbara region and other areas in Western Australia, Queensland, New South Wales, Tasmania and South Australia. Although the studies are not extensive, the availability of void spaces in the aquifer, and recharge of oxygenated water and organic matter appeared to be consistent determinants

The Investigation Area related to RAAF Base Darwin has a shallow unconfined aquifer, which receives fresh infiltration in the Wet season, and would likely provide sufficient oxygen and organic matter to sustain stygofauna. Groundwater is present in a shallow honeycomb ferricrete and lateritic clays, which present as gravels in some areas, in the Wet season and in the underlying low permeability weathered siltstone. As groundwater levels drop in the Dry season, the voids in the ferricrete become dry and the aquifer would not represent a permanent habitat for stygofauna communities. Therefore, stygofauna are considered unlikely to be present in the RAAF Base Darwin Investigation Area and have not been specifically addressed further in this ERA.

3.3. Terrestrial Biota Mammals within the investigation area include dingo (Canis lupus dingo), northern brown bandicoot (Isoodon macrourus), wallabies (Macropus agilis), flying fox (aka fruit bats) (Pteropus spp.), microbats (such as the common sheath-tailed bat Taphozous georgianus), and other small mammals such as rodents, gliders, and possums. Northern quoll (Dasyurus hallucatus) also inhabit the area but are rare due to the impact of the cane toad. Bird species occurring in the investigation area include birds of prey such as eagles and kites, forest birds such as fruit dove and fig bird, woodland birds, and aquatic birds such as cormorants, darters, egrets, jacana, and jabiru. In addition to the saltwater and freshwater crocodiles discussed above, there are a variety of other reptiles and amphibians within the investigation area, including goanna, skinks, geckos, water snakes, turtles, and frogs. The cane toad (Rhinella marina) is a problematic introduced species as it is toxic to important native species including quolls and goannas.

Table 3-6 lists the terrestrial vertebrates that were observed and collected for PFAS analysis during the assessment.



40

Table 3-6: Summary of amphibian, reptile, and mammal species sampled and collected around Rapid Creek and the Darwin investigation area by Coffey in March 2018 Amphibian Common Name (Species Name)

Reptile Common Name (Species Name)

Mammals and Birds Common Name (Species Name)

Cane toad (Rhinella marina) Red-sided Rainbow Skink (Carlia rufilatus)

Southern Myotis (Myotis macropus)

Northern Dwarf Tree-Frog (Litoria bicolor)

Burton’s Legless Lizard (Lialis burtonis)

Grassland Melomys (Melomys burtoni)

Ornate Burrowing Frog (Platyplectrum ornatus)

Striped Rainbow Skink (Carlia munda)

Northern Brown Bandicoot (Isoodon macrourus)

Marbled Frog (Limnodynastes convexiusculus)

Two-Lined Dragon (Diporiphora bilineata)

Brushtail Possum (Trichosurus vulpecula)

Green Tree-Frog (Litoria caerulea) Metallic Snake-eyed Skink (Cryptoblepharus metallicus)

Magpie Goose (Anseranas semipalmata)

Zig-zag Gecko (Oedura rhombifer) Bynoe’s Gecko (Heteronotia binoei)

Green tree frog (Litoria australis) Asian House Gecko (Hemidactylus frenatus)

Rocket Frog (Litoria nasuta) Spotted Tree Monitor (Varanus scalaris)

Keelback Snake (Tropidonophis mairii)

Children’s Python (Antaresia childreni)

Northern Spotted Rock Dtella (Gehyra australis)

Port Essington Ctenotus (Ctentous essingtonii)

Slaty-grey Snake (Stegonotus cucullatus)

Gilbert’s Dragon (Lophognathus gilberti)

Northern Small-eyed Snake (Cryptophis pallidiceps)

Terrestrial invertebrates also form an important part of the food web and may accumulate PFAS through contact with contaminated streams, consumption of contaminated plants or burrowing in soil. Table 3-7 lists the invertebrates observed and collected for PFAS analysis during the investigation. Collected species were grouped for analysis based on behaviour (ground dwelling or aerial) and location caught.



41

Table 3-7: Summary of terrestrial invertebrates collected in December 2017 Invertebrate groups from Rapid Creek

Invertebrate groups from Ludmilla Creek

Invertebrate groups from On-Base

Beetles, cicadas, cockroaches Beetles, cicadas, cockroaches Beetles, cicadas, cockroaches

Grasshoppers, crickets, praying mantis



Moths, flies, dragonfly and aerial species



Ants, caterpillars, spiders, worms, bugs



Green ants

3.4. Receptors of Concern (ROC) It is not feasible to evaluate the potential risks to all the wildlife species that may occur within the investigation area. Therefore, a subset of species (indicator species) was selected for evaluation in the ERA that are considered representative of the occupied habitats and trophic levels (e.g., herbivores, carnivores) in the investigation area. The species selected are also considered to be representative of the environmental values to be protected and of the most sensitive and potentially highly exposed species.

Below is a list of the trophic levels and species that were evaluated as ROCs for the aquatic and terrestrial environments in the investigation area. Multiple factors were considered in developing the appropriate ROCs to be evaluated including diet, habitat, and whether the species has been sighted on the Base including during the habitat surveys. The results of the habitat survey were used to finalise the list of ROCs that were evaluated in the ERA. Below is a list of the generic trophic levels that are to be protected and Table 3-8 includes the ROCs determined to be evaluated for these trophic levels (i.e. specific indicatory species).

Aquatic Environment (including Rapid Creek, Ludmilla Creek, Reichardt Creek and Sadgroves Creek):

• Aquatic Plants

• Aquatic Invertebrates

• Aquatic Vertebrates (i.e., fish)

• Invertivorous and Omnivorous Birds (i.e., primarily feeding on aquatic invertebrates; e.g., jacana)

• Piscivorous Birds (i.e., primarily feeding on fish; e.g., egrets, cormorants, jabiru)

• Piscivorous Mammals (e.g., water rat)

• Piscivorous Reptiles (e.g., water snakes)

Terrestrial Environment:

• Terrestrial Plants

• Terrestrial Invertebrates

• Herbivorous Birds (i.e., primarily feeding on plant matter; e.g., fruit dove, fig bird)

• Herbivorous Mammals (e.g., flying fox)

• Invertivorous and Omnivorous Birds (i.e., primarily feeding on terrestrial/soil invertebrates; e.g., woodland birds)

• Invertivorous and Omnivorous Mammals (e.g., rodents)



42

• Invertivorous and Omnivorous Reptiles (e.g., skinks, gecko)

• Carnivorous Birds (i.e., primarily feeding on other vertebrate animals; e.g., eagles, kites)

• Carnivorous Mammals (e.g., quoll)

• Carnivorous Reptiles (e.g., goanna)

ROCs have been selected from the species that may be present on the Base or in the surrounding areas, based on information provided by:

• Limited habitat surveys which were performed on Base and in the surrounding areas during site visits and in the course of sample collection.

• Commonwealth Department of the Environment (DoE) Protected Matters Search Tool, to identify ecological Matters of National Environmental Significance (MNES) within 20 km of the Base.

• Living Atlas of Australia.

• Plant and animal species listed under the Northern Territory Territory Parks and Wildlife Conservation Act of 2001.

• The Preliminary Master Plan for Darwin International Airport (DIA, 2017).

• Ecological Assessment for F-35 aircraft operations (ELA, 2014).

The prime factors that were considered in the selection of potential ROCs to be evaluated include diet, body weight, habitat, and presence at the site. Details surrounding these factors are summarised below:

• Diet – As part of the risk assessment process, the potential for uptake of constituents through food sourced at the site is evaluated. Therefore, assessing receptors that 1) have a diet of prey that may have accumulated PFAS and 2) cover different feeding guilds including herbivores, omnivores, piscivores and carnivores.

• Body weight – The body weight of receptors is directly correlated to food ingestion rate (Nagy 2001) and food web modelling is sensitive to body weight. Therefore, selecting ROCs with varying body weights ensure that risks to different sized organisms are characterized.

• Habitat – Because the IA has diverse types of habitats, by choosing ROCs that may inhabit a variety of habitat can ensure that species in the IA are adequately characterized.

• Sightings – The ROCs chosen are those that have previous been sighted in the IA or where noted to be in the area during abiotic and biotic media sampling.

Details of these primary factors for each of the selected ROCs are provided in Appendix M.



43

Table 3-8: Receptors of concern and ecological exposure pathways considered in food web modelling ROC

(Common Name)

Exposure Pathway for PFOS/PFOA Notes

Ingestion (Diet)

Ingestion (Surface Water)

Incidental Ingestion

(Soil/Sediment)

Invertivorous and Omnivorous Aquatic Birds

White-Faced Heron Yes Yes Yes Some incidental ingestion of sediment is assumed for all species within this trophic group based on the food items and foraging nature of the selected species.

Comb-Crested Jacana Yes Yes Yes

Eastern Great Egret Yes Yes Yes

Pacific Black Duck Yes Yes Yes

Radjah Shelduck Yes Yes Yes

Piscivorous Aquatic Birds

Black-Necked Stork Yes Yes Yes The black-necked stork feeds on fish, small crustaceans, and amphibians and is likely to incidentally ingest sediment.

Little Pied Cormorant Yes Yes Yes The little pied cormorant feeds on fish, crustaceans, and plant material and is likely to incidentally ingest sediment.

Little Black Cormorant Yes Yes No The little black cormorant feeds primarily on fish and it is not considered likely to ingest sediment,

Australasian Darter Yes Yes No Australasian darter diet consists mainly of fish and it is not considered likely to ingest sediment.

Piscivorous Mammals

Large-Footed Myotis Yes Yes No The large-footed myotis consumes aquatic insects and fish and is not considered likely to ingest sediment.

Piscivorous Reptiles

Gray’s Water Snake Yes Yes Yes Gray’s water snake is a burrowing snake that feeds on gobies, crustaceans and nudibranchs and is considered likely to incidentally ingest soil / sediment.

Keelback Snake Yes Yes Yes Keelback snake feeds mainly on frogs but also eats fish, reptile eggs, and mammals and is considered likely to incidentally ingest soil / sediment.

Northern Snake-Necked Turtle

Yes Yes Yes With a diet of insects, fish, crustaceans, tadpoles, frogs, and carrion, the Northern Snake-necked turtle is considered likely to incidentally ingest sediment.



44

ROC

(Common Name)


Ingestion (Diet)



(Soil/Sediment)

Herbivorous, Invertivorous and Omnivorous Aquatic Reptiles

Northern Snapping Turtle Yes Yes Yes Based on a diet consisting of a majority of fruit and leaves of riparian trees, the northern snapping turtle may only incidentally ingest sediment.

Northern Yellow-Faced Turtle

Yes Yes Yes Although the exact diet is unknown, it is likely that the northern yellow-faced turtle’s diet consists of mussels, molluscs, snails, and vegetation, which makes it likely to incidentally ingest sediment.

Herbivorous Terrestrial Birds

Rose-Crowned Fruit Dove

Yes Yes Yes Rose-crowned fruit dove’s diet consists mainly of seeds and may only incidentally ingest minor of amounts of soil associated with seeds that have fallen on the ground.

Australasian Figbird Yes Yes No Australasian diet consists mainly of figs and other soft fruits as well as insects and may only incidentally ingest minor of amounts of soil

Magpie Goose Yes Yes Yes With a diet that consists of seeds, grasses, and rhizomes these species may incidentally ingest soils. Bar-Shouldered Dove Yes Yes Yes

Plumed Whistling Duck Yes Yes Yes

Herbivorous Terrestrial Mammals

Pale Field Rat Yes Yes Yes The pale field rats diet consists of roots, grass stems, and seeds, therefore it is likely to incidentally ingest soil.

Agile Wallaby Yes Yes Yes Some incidental ingestion of soil is assumed during grazing.

Black Flying Fox Yes Yes No Species diet consists mainly of fruits, pollen, and nectar and it is not considered likely to ingest soil.

Black-Footed Tree Rat Yes Yes No The diet of the black-footed tree rat consists of fruit, seeds, insects, flowers, nectar, and the green tips of vegetation, it may incidentally ingest minor amounts of soil.

Grassland Melomys Yes Yes No Grassland melomys diet is primarily grass vegetation, seeds, cane fiber and other vegetation, thus it may incidentally ingest minor amounts of soil

Invertivorous/Omnivorous Terrestrial Birds

Masked Lapwing Yes Yes Yes Based on diet some incidental ingestion of soil is assumed for both of these species.

Straw-Necked Ibis Yes Yes Yes



45

ROC

(Common Name)


Ingestion (Diet)



(Soil/Sediment)

Black-Faced Wood swallow

Yes Yes No The diet of the Black-faced Wood swallow consists of insects, and it will also eat nectar, thus incidental ingestion of soil is not expected.

Invertivorous/Omnivorous Terrestrial Mammals

Short-Beaked Echidna Yes Yes Yes The short-beaked echidna feeds primarily on ants, termites, grubs and worms and is likely to incidentally ingest soils.

Common Rock Rat Yes Yes Yes Some incidental ingestion of soil is assumed for this species based on the food items and foraging nature of the selected species.

Northern Brown Bandicoot

Yes Yes Yes Some incidental ingestion of soil is assumed for this species based on the food items and foraging nature of the selected species.

Invertivorous/Omnivorous Terrestrial Mammals (Continued)

Large Bent-Winged Bat Yes Yes No Based on the preferred diet, incidental ingest of soil by these species is unlikely.

Orange Leaf-Nosed Bat Yes Yes No

Invertivorous/Omnivorous Terrestrial Reptiles

Bynoe’s Gecko Yes Yes No Based on the preferred diet there is some potential for incidental ingest of soil by these species. Gilbert’s Dragon Yes Yes No

Striped Rainbow Skink Yes Yes No

Carnivorous Terrestrial Birds

Black-Shouldered Kite Yes Yes No No incidental ingestion of soil is assumed for these species due to the nature of hunting (swooping).

It is acknowledged that carnivorous terrestrial birds typically ingest very little water, however for the purpose of this ERA the water ingestion pathway has been conservatively considered to be complete.

Wedge-Tailed Eagle Yes Yes No

Whistling Kite Yes Yes No

Black Kite Yes Yes No

Carnivorous Terrestrial Mammals

Northern Quoll Yes Yes Yes The Northern Quoll is known to feed on a variety of prey including invertebrates found in leaf litter. Therefore, incidental ingestion of soil is assumed to occur.



46

ROC

(Common Name)


Ingestion (Diet)



(Soil/Sediment)

Dingo Yes Yes Yes Dingo diet consist mainly of small mammals but may consume insects and birds. Due to the nature of consuming food on the ground, the incidental ingestion of soil is assumed.

Carnivorous Terrestrial Reptiles

Floodplain Monitor (Yellow-Spotted Monitor)

Yes Yes No Monitors are known to consume insects and turtle eggs, thus incidental ingestion of soils is not considered likely to occur.

Olive Python Yes Yes No The diet of the olive python includes mammals, birds, bats, and other reptiles, thus incidental ingestion of soil is not considered likely to occur.

Brown Tree Snake Yes Yes No As the diet of the brown tree snake consists primarily of small birds, amphibians, lizards and other reptiles, it is not considered likely that this species incidentally ingests soil.



47

3.5. Potential Complete Exposure Pathways If a receptor is to be exposed to a chemical contaminant deriving from the site, a complete exposure pathway must exist. An exposure pathway describes the course some chemicals take from a source to the exposed receptors and generally includes the following elements:

• a source and mechanism of chemical release;

• a retention or transport medium (or media where chemicals are transformed between media);

• a point of potential contact between the receptor and the contaminated media (it should be noted that a contaminated medium is considered to be one in which chemicals of potential concern are reported above adopted relevant screening criteria); and

• an exposure route (e.g., ingestion) at the point of exposure.

3.5.1. Complete Exposure Pathways The following exposure pathways are considered likely to be complete for terrestrial organisms at the RAAF Darwin DIA:

• incidental ingestion of PFAS adhered to soil (animals – invertebrates and vertebrates);

• uptake and bioaccumulation of PFAS from soil moisture (plants and soft-bodied soil invertebrates);

• ingestion of dissolved PFAS in freshwater sources (animals – invertebrate and vertebrate); and

• ingestion of PFAS bioaccumulated in food sources (animals – plants, invertebrates and vertebrates).

Uptake and bioaccumulation of PFAS from the environment into plants has been assessed through direct measurement of PFAS in plant samples. Therefore, this exposure pathway has not been included in food web modelling. The ingestion of contaminated plant material has been included in the food web modelling.

The following exposure pathways are considered likely to be complete for aquatic (freshwater and estuarine) organisms:

• incidental ingestion of PFAS adhered to sediments (animals – invertebrates and vertebrates);

• uptake and bioaccumulation of PFAS from sediment pore water and/or surface water (all aquatic organisms);

• ingestion of PFAS bioaccumulated in food sources (animals – aquatic plants, invertebrates and vertebrates); and

• uptake and bioaccumulation of PFAS from surface water into aquatic plants including plants within the riparian zone.

3.5.2. Insignificant Exposure Pathways The following pathways are considered to be insignificant for ecological receptors at the Base and in surrounding assessment areas;

• Dermal uptake of PFAS from environmental media – dermal uptake is a minor pathway for PFAS uptake based on current understanding of the ability for it to cross into/through dermal membranes.



48

• Ingestion of PFAS in groundwater – it has been assumed that surface water samples collected to date (which include samples from areas where groundwater has been observed to discharge to ground surface) provide a better representation of potential exposures because of water ingestion.

• Direct contact with groundwater at the Base and in surrounding assessment areas has not been considered as direct sampling of surface water has been undertaken in the various waterways which are connected to the Base. It is understood that there is potential for groundwater seeps to occur in areas around the Base; however, these are most likely to occur following periods of heavy rainfall, and therefore dilution of reported groundwater concentration is considered likely to occur. Assessment of potential toxicity associated with direct consumption of reported PFAS concentration in surface water is considered appropriate, as potential exposure to groundwater is likely to be intermittent.

• Direct contact with soil beyond 0.5 m bgl at the Base and in the surrounding assessment areas has not been considered as a significant exposure pathway. Only shallow soil, 0.0 – 0.5 m bgl, have been evaluated because this is considered the most biologically active zone and where ecological receptors would likely be exposed (US EPA 2015).

• Direct contact with stygofauna in groundwater at the Base and in surrounding assessment areas. A review identified that the aquifer at the investigation area is unlikely to provide a permanent habitat for stygofauna communities, as such stygofauna are considered unlikely to be present in the investigation area.

3.5.3. Complete Exposure Pathways Considered in Food Web Modelling

Where one or more of the elements listed in Section 3.5 is missing, the exposure pathway is considered to be incomplete and there is therefore no risk to the receptor. Ecological exposure pathways and the extent to which they have been considered as complete as part of the food web modelling undertaken for selected ROC, have been provided in Table 3-8. Exposure pathways have been determined based on the diet and behaviour of the ROC.

Rapid Creek, Ludmilla Creek and Darwin Harbour (including Sadgroves Creek and Reichardt Creek) are considered the main water supplies for organisms in the investigation area, and have been reported to contain detectable concentrations of PFAS in surface water and/or sediment. For a conservative assessment, the exposure point concentration derivation was dependent on the number of sampling locations and the number of detected concentrations. For those EEAs with a robust data set, the 95% UCL was used to avoid one value skewing the maximum concentration. The full data sets are presented in Appendix G and the statistical assessment in Appendix H. Terrestrial soils surrounding potential source areas in the investigation area have been reported to contain detectable concentrations of PFAS. As noted previously, Section 4.2 outlines the rules used when evaluating the data set and deriving the relevant exposure point concentration for each terrestrial EEA.



49

4. Exposure Assessment Exposure assessment involves compiling information about habitat at the Base, receptor diet, potentially complete exposure pathways, exposure point concentrations, exposure duration and estimated chemical intakes to assess the likely magnitude of exposure (ASC NEPM, 2013, ANZECC, 2000).

4.1. Exposure Point Concentration Selection An Exposure Point Concentration (EPC) is the assumed concentration of contaminant in the medium to which the population is exposed, at the point of exposure (ASC NEPM). In accordance with ASC NEMP guidelines EPCs were estimated through statistical analysis of site-specific data, obtained from the environmental investigations undertaken on- and off-Base as summarised in Appendix G. Summary media concentration ranges are presented in sampling areas on Figure 12, Appendix A.

4.1.1. Data Analysis

All abiotic and biotic data was analysed using the software program ProUCL (US EPA Version 5.1) to obtain the distribution, 95% percentile Upper Confidence Limit (UCL) based on the appropriate distribution, average, standard deviation and proportion of detects and non-detects in the samples where appropriate. ProUCL outputs are presented in Appendix H. These parameters were interpreted using the methodology described in Table 4-1 to estimate an appropriate EPC for each area of ecological concern. Data sets with fewer than nine samples were considered to be insufficient for statistical analysis and the maximum concentration was adopted as a conservative estimate of the EPC.

All statistical analyses were undertaken excluding non-detects to provide a conservative estimate of the average and 95% UCL concentrations. This approach is considered conservative as it is likely to provide a higher estimate of the average and 95% UCL concentrations. Inclusion of non-detects at either the full detection limit or half the detection limit would increase the number of data points in the lower end of the data range and therefore reduce the estimated average or 95% UCL concentration obtained. The exclusion of non-detects from the data set is considered appropriate for this level of assessment (i.e. preliminary ERA) as it accounts for habitats that are in close proximity to a source area, rather than assuming a home range that extends evenly across the whole EEA.



50

Table 4-1: Decision factors used to determine EPCs for abiotic and biotic media

Decision Factors EPC Source Type

Rationale

< 9 samples and < 9 detects Maximum reported concentration

The maximum reported concentration for on- and off-Base was adopted where there was insufficient data to conduct statistical analysis.

≥9 samples and ≥ 9 detects and ≥ 50% detects (expressed as proportion of detects)

95% UCL The 95% UCL was considered to be an appropriately conservative representation of data (as per recommendations in the ASC NEPM), provided there were sufficient data points and the data distribution was not significantly negatively skewed. Professional judgement has been applied and nine or more data points have been considered to be sufficient. It is noted that the ProUCL User’s Guide (US EPA, 2013) indicates that a sample size ≤6 is considered not to produce a reliable statistical evaluation for risk assessment and remediation purposes.

≥ 9 samples and ≥ 9 detects and < 50% detects (expressed as proportion of detects)

Or

≥ 9 samples and ≤ 9 detects

Average detected concentration

For highly skewed data, particularly negatively skewed data, the 95% UCL is unreliable. The arithmetic average was calculated (excluding non-detects) as appropriate using ProUCL.

Non-Detect Laboratory Limit of reporting

Where PFOS and PFOA were not detected in a sampled environmental medium, the maximum laboratory limit of reporting was adopted as an exposure point concentration in the food web modelling undertaken.

Detect = sample where the analyte was reported at a concentration above the laboratory limit of reporting.



51

4.1.2. Summary of Adopted Exposure Point Concentration The exposure point concentrations (EPCs) selected for use in the ecological risk assessment represent the maximum detected value, the 95% UCL, the average detected value or the limit of reporting in each abiotic or biotic media. This is conservative and any constituents that are eliminated based on this can be assumed to not pose a potential risk, but those that are not eliminated still may not pose a potential risk based on a more appropriate exposure point concentration. Therefore, for this preliminary ecological risk assessment the use of conservative values is appropriate.

The data analysis to select EPCs in soil only considered the soil zone in which most biological activity is thought to occur, which is between 0.0 and 0.5 m below ground level (bgl). This soil zone accounts for activity by burrowing organisms (invertebrate and vertebrate) and plants.

Porewater was not analysed for PFOS and PFOA concentrations, therefore, food web modelling has been undertaken based on surface water concentrations only, and no EPC have been reported for porewater.

Mammals exposure point concentrations were developed from whole body, organ only, flesh only and whole body based on blood serum samples. Whole body, organ only, and flesh only samples were used as reported in µg/kg, but the blood serum concentrations reported in ng/mL were extrapolated to tissue concentration using the animal tissue to serum ratios from ToxConsult (2016, as cited in AECOM 2017). Average tissue to serum ratios from the referenced studies were applied. PFOS tissue to serum ratios in the liver and kidney were 7.15 and 0.6, respectively, and for PFOA were 1.34 and 0.45, respectively. Using body, serum and organ weights (IFFA CREDO Laboratory Animals, 2018), the whole body to serum ratio was calculated. For PFOS, the whole body to serum ratio was 0.27 and for PFOA the whole body to serum ratio was 0.09. By using these ratios, the measured blood serum concentration was converted to whole body tissue concentration. This value may underestimate the specific tissue concentrations where organs accumulate PFOS and PFOA, however inclusion of the contribution of serum is expected to reasonably estimate the whole body concentration and account for most organs and muscle, as serum would be distributed through these body parts. Appendix K contains a detail analysis of the evaluation of blood serum for the derivation of whole body tissue concentrations.

The exposure point concentrations for reptiles and amphibians have been developed based on whole organisms and composite samples. For the food web modelling the calculated EPCs for reptiles and amphibians have not been adjusted to exclude data where reptiles and amphibians are a food source for reptile and/or amphibian receptors of concern.

Where PFOS and PFOA were not detected in abiotic or biotic media, it has therefore been assumed that concentrations may be equal to the laboratory reporting limit. Therefore, the laboratory LOR has been adopted as the EPC.

Table 4-2 and 4-3 provides a summary of the EPC adopted for the food web modelling for each EEA. Tables 1 through 12 in Appendix G provide full sample results and a detailed summary of the statistical analysis completed for all exposure point concentrations is presented in Appendix H.



52

Table 4-2: Summary of adopted PFOS/PFOA exposure point concentrations – aquatic food web modelling (Off-Base) Catchment Ecological Exposure Area Sediment Surface

Water Aquatic Plants

Aquatic Invertebrates (a)

Fish (a) Reptiles & Amphibians

Mammals

µg/kg - DW µg/L µg/kg - WW µg/kg - WW µg/kg - WW µg/kg – WW µg/kg - WW

PFOS

Rapid Creek Rapid Creek Freshwater (Off-Base) 182.3 (c) 0.71 (c) 13 (d) 78 (d) 764 (c) 2262 (c, h) 183 (c, h)

Rapid Creek Estuary (Off-Base) 19 (d) 0.30 (d) 2.2 (d) 19.4 (b) 29 (c) 20.7 (c, i) 5.0 (c, i)

Ludmilla Creek Ludmilla Creek (Off-Base) 21 (d) 3.0 (c) 4 (d) 14.1 (b) 51.4 (c) 31.61 (c, g) 36.7 (c, g)

Reichardt and Sadgroves creeks

Reichardt & Sadgroves Estuary (Off-Base) 43.6 (b) 0.10 (c) 0.7 (d) 3.1 (b) 1.7 (c) 24 (d, j) NM

All Estuarine/Marine 24.4 (b) 1.7 (c) 1.3 (b) 14.4 (b) 25.8 (c) 17.78 (c) 36.7 (c)

PFOA

Rapid Creek Rapid Creek Freshwater (Off-Base) <5 (e) 0.02 (c) <0.5 (e) 1.3 (d) <0.5 (e) <0.5 (e, h) 0.09 (d, h)

Rapid Creek Estuary (Off-Base) <5 (e) 0.01 (d) <0.5 (e) 2.2 (b) <0.5 (e) <0.5 (e, i) <0.09 (e, i)

Ludmilla Creek Ludmilla Creek (Off-Base) <5 (e) 0.07 (b) <0.5 (e) 4.4 (b) 0.34 (b) <0.5 (e, g) 0.52 (d, g)

Reichardt & Sadgroves creeks

Reichardt & Sadgroves Estuary (Off-Base) <5 (e) 0.01 (b) <0.5 (e) 1.7 (b) <0.5 (e) <0.5 (e, j) NM

All Estuarine/Marine <5 (e) 0.05 (b) <0.5 (e) 3.6 (b) <0.5 (e) <0.5 (e) 0.52 (d, h) Notes: DW = dry weight (concentration reported based on dry weight) WW = wet weight (concentration reported based on wet weight) NM = Not measured a – Fish and aquatic invertebrates were only collected during the wet season and these tissue concentrations will be used for both wet and dry season analysis. b – Calculated average of all detected concentrations. c – Calculated 95%UCL using ProUCL 5.1 software. d – Maximum detected concentration. e – Level of reporting – where no detectable concentrations were reported it has been assumed that the contaminant may be present at concentrations up to the level of reporting. f – Soil measured in Zone C represents aquatic sediment from drains in Zone C. g – Reptiles, Amphibians, and Mammals captured in Zone C soil zone. h – Reptiles, Amphibians, and Mammals captured in Zone A soil zone. i – Reptiles, Amphibians, and Mammals captured in Zone B soil zone. j – Reptiles, Amphibians, and Mammals captured in Zone D soil zone.



53

Table 4-3: Summary of adopted exposure point concentrations – terrestrial food web modelling (On-Base) Ecological Exposure Area

Surface Soil

(0 – 0.5 m bgl)

Surface Water Terrestrial Plants Terrestrial Invertebrates

Fish(a) Reptiles & Amphibians

Mammals Birds

µg/kg - DW µg/L µg/kg - WW µg/kg - WW µg/kg - WW µg/kg – WW µg/kg - WW µg/kg - WW

PFOS

Zone A 18.8 (b) 2.16 (c, f) 7.1 (b) 250 (d) 764 (c, i) 2,262 (c) 183 (c) 7,200 (d, g)

Zone B 2,999 (c) 1.43 (c, f) 2.9 (b) 99.9 (c) 29.1 (c, h) 20.7 (c) 5.0 (d) 17 (d, g)

Zone C 1,590 (c) 0.43 (c, f) 5.0 (b) 15 (d) 51.4 (c, j) 31.6 (c) 36.7 (d) 69 (d, g)

Zone D 3,223 (c) 0.73 (c, f) <0.3 (e) 3.8 (d) 1.73 (c, k) 24 (d) NM (L) (85.2 (c)) 7.9 (d, g)

PFOA

Zone A <0.5 (e) 0.06 (c, f) <0.5 (e) <0.5 (e) <0.5 (e, i) <0.5 (e) 0.09 (d) <0.5 (e, g)

Zone B 29.2 (b) 0.24 (c, f) 1.2 (b) 0.9 (d) <0.5 (e, h) <0.5 (e) <0.09 (e) <0.5 (e, g)

Zone C 47.3 (b) 0.02 (b, f) <0.5 (e) <1.0 (e) 0.34 (b, j) <0.5 (e) 0.52 (d) <0.5 (e, g)

Zone D 25.8 (b) 0.02 (b, f) <0.5 (e) 4.5 (e) <0.5 (e, k) <0.5 (e) NM (L) (0.52 (d)) <0.5 (e, g) DW = dry weight (concentration reported based on dry weight) WW = wet weight (concentration reported based on wet weight) NM = Not measured a – Fish tissue data relates to nearest waterway relevant to the terrestrial zone. b – Calculated average of all detected concentrations. c – Calculated 95%UCL using ProUCL 5.1 software. d – Maximum detected concentration. e – Level of reporting – where no detectable concentrations were reported it has been assumed that the contaminant may be present at concentrations up to the level of reporting. f – Surface water values used for the evaluation of terrestrial organisms was from the drains in each of the soil zones (Zone A, Zone B, Zone C, Zone D). g –Bird tissue was not measured and Reptile tissue data was used as a surrogate. h – Fish tissue collected from the Rapid Creek Estuary site was used to evaluate the Zone B EEA. i – Fish tissue collected from the Rapid Creek Freshwater site was used to evaluate the Zone A EEA. j– Fish tissue collected from the Ludmilla Creek site was used to evaluate the Zone C EEA. k – Fish tissue collected from the Darwin Harbor including Sadgroves and Reichardt creeks site was used to evaluate Zone D EEA L – In the absence of Zone specific mammal data for Zone D, the mammal data from all areas was considered.



54

4.2. Other PFAS compounds Exposure point concentrations for PFOS and PFOA have been adopted for this ERA to enable an assessment of risks associated with these two primary risk driving PFAS compounds. However, it is recognised that ‘other’ PFAS compounds were reported in the environment and that the potential increase in toxicity from these other PFAS compounds may need to be considered in the ERA.

A range of ‘other’ PFAS compounds have been reported in abiotic and biotic media including PFHxS, PFHxA, PFPeA, PFPeS, PFBS, PFNA, 6:2 FtS and 8:2 FtS. The relative contribution of the other PFAS compounds to the total PFAS concentrations is shown in Figure 4-1 to Figure 4-5 for abiotic and biotic media.

Figure 4-1: Percentage of detected PFAS compounds in soil

PFOS was the dominant PFAS compound detected in soil in each zone and represented greater than 90% of total PFAS in Zones A and D, and greater than 80% of total PFAS in Zones B and C. PHFxS and PFHxA were the generally the next most dominant PFAS compounds detected in soil. Multiple compounds were reported in Zone B around the current fire training area, and the relative concentrations are shown in Figure 4-2. The chart shows that 88% of the reported concentrations were PFOS+PFHxS, with the next most prominent compounds being PFUnDA, PFDS, PFTriDA and PFOSA. Other compounds represented less than 1% each.



55

Figure 4-2: Percentage of PFAS compounds in soil from Zone B

Figure 4-3: Percentage of detected PFAS compounds in surface water



56

In surface waters PFOS was the most dominant PFAS compound detected in each area, generally representing 50 to 70% of total PFAS. The next most dominant PFAS compound detected in surface waters was PFHxS. In each waterway PFOS and PFHxS constitute approximately 85% of the total PFAS concentrations. The exceptions being drains in Zone B, which includes run-off from recent or current fire fighting services, and Reichardt Creek. PFHxA, PFPeS and PFBS were the next most prevalent across all water ways, with PFOA, PFHpA and PFPeA contributing at least 10% to drains in Zone B and Reichardt Creek.

Figure 4-4: Percentage of detected PFAS compounds in terrestrial biota

The dominant PFAS compound detected in all terrestrial biota groups was PFOS (> 90%), with the exception of plants, which comprised approximately 12% PFOS and PFBA (60%) being the dominant compound. The average plant composition is dominated by a single sample result, and when that result (a sample of Poinciana leaves from the current fire training area) is removed PFOS represents approximately 50% of the total, with PFHxS being the next most dominant compound (30%).



57

Figure 4-5: Percentage of detected PFAS compounds in aquatic biota

PFOS was the dominant PFAS compound detected in all aquatic biota groups, representing greater than 90% of total PFAS in fish and estuarine aquatic plants, and greater than 70% in freshwater aquatic plants and estuarine invertebrates. PFOS represented a lower proportion of total PFAS (50%) in freshwater invertebrates, in this case PFHxS was the next most dominant compound (30%). PFHxS was also the second most dominant PFAS compound in the other aquatic biota.

Table 4-4: Percentage of total measured PFAS that was not PFOS or PFOA

Elements of the Food Web Approximate percentage of PFAS compounds not including PFOS and PFOA

Soil <15%

Terrestrial plants 50%

Amphibians and Reptiles <10%

Mammals <10%

Invertebrates <10%

Surface Water 50%

Aquatic plants <25%

Aquatic Invertebrates 50%

Fish 10%

There is very little toxicity data available on other PFAS compounds, and therefore weighted toxicity impact has not been calculated. Based on the relative percentages of PFOS to total reported PFAS compounds, there is a recognised potential that the specific pathways of ingestion of terrestrial plants, surface waters and aquatic invertebrates may underestimate total risk from PFAS, by considering PFOS and PFOA alone. Other diet items did not contain significant quantities of other PFAS compounds.



58

4.3. Food Web Modelling Due to the structure and chemical properties of PFAS, the estimation of bioaccumulation using traditional methods (i.e. application of bioaccumulation factors) is not considered appropriate to model concentrations in biotic media (i.e., plants, invertebrates, etc) that are part of the food web. Therefore, quantitative food web modelling, in accordance with methods provided by the US EPA (1993 and 1999) and Sample et al (1996 and 1997), has been undertaken to evaluate the potential exposure to terrestrial and aquatic receptors including mammals, birds, and reptiles. Food web modelling involves the estimation of intake rates for PFAS using available environmental data, measured prey tissue data and assumed receptor-specific exposure parameters (i.e., body weight, diet composition, etc).

PFOS and PFOA doses were calculated for each ROC using the intake calculations presented in Section 4.3.1, using the relevant wildlife exposure parameters discussed in Section 4.3.2. The PFOA and PFOS doses for each ROC using the calculations in Section 4.3 are presented in Appendix N.

4.3.1. Intake Calculations Species-specific intakes of contaminants (PFOS and PFOA) accumulated in food sources, adhered to soil or sediment, or dissolved in water have been estimated using the following equations (adopted from Sample et al (1997)):

𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝐼𝑗𝑗 = 𝑃𝑃ℎ (𝐴𝐴𝐻𝐻𝐻𝐻

[��𝑃𝑃𝑖𝑖𝑖𝑖(𝐼𝐼𝑖𝑖 𝑥𝑥 𝐶𝐶𝑗𝑗𝑖𝑖)])𝑛𝑛

𝑖𝑖=1

𝑚𝑚

𝑖𝑖=1

Where:

Intakej = total oral exposure (intake) to contaminants (j) (mg/kg/day) (intake rate)

Ph = proportion of suitable habitat in the contaminated area

A = area contaminated (ha)

HR = home range size (ha) of ROC

m = total number of ingested media (e.g. food, water, or soil)

n = number of types of media (i) consumed (unitless)

Pik = proportion of type (k) of medium (i) consumed (unitless)

Ii = ingestion rate of medium (i) (kg/kg body weight/day or L/kg body weight/day)

Cji = concentration of contaminant (j) in medium (i) (mg/kg or mg/L)

It is noted that the proportion of suitable habitat in the impacted area, area impacted and home range size have not been applied for any species which would not spend all their time at the Base or in the surrounding IA. This is considered a conservative approach as it indicates that migratory/nomadic species would spend all their time in the Base or surrounding investigation area. This is discussed further in the Uncertainties section.

𝐼𝐼𝑖𝑖 = 𝐼𝐼𝐼𝐼𝐼𝐼𝐻𝐻𝑖𝑖/𝐵𝐵𝐵𝐵

Where:

Ii = ingestion rate (kg/kg body weight/day or L/kg body weight/day)

IngRi = ingestion rate of media (i) (soil/sediment/water or food source) (kg/day or L/day)

BW = body weight of the organisms for which intake is being calculated (kg)



59

Food ingestion rates for Australian species are not readily available in the literature. However, ingestion rates for a selection of Australian native species are provided in Nagy (2001). Therefore, where available, the estimated food ingestion rates based on field studies has been adopted from Nagy (2001) (Tables 4-5 and 4-6). Otherwise food ingestion rates for each species have been estimated based on allometric equations published in US EPA (1993) and Nagy (2001). These equations are based on research conducted by Nagy (2001) where the field metabolic rates of 79 species of mammals (including marsupials) and 95 bird species were measured. The food ingestion rate calculation is as follows:

𝐼𝐼𝐼𝐼𝐼𝐼𝐻𝐻𝑖𝑖𝑖𝑖𝑖𝑖 = 𝐼𝐼𝐵𝐵𝐼𝐼𝑏𝑏

Where:

IngRiWW = food ingestion rate (g WW/day)

Wt = body weight of receptor species (g WW)

a = empirical coefficient (all birds = 2.065; all mammals = 0.794; all reptiles = 0.0333) (adopted from Nagy 2001)

b = empirical coefficient (all birds = 0.689; all mammals = 0.773; all reptiles = 0.932) (adopted from Nagy 2001)

Equations presented in USEPA (1999) were used to estimate water ingestion rates for each ROC (birds and mammals). USEPA (1999) does not present an equation for the water ingestion rate of reptiles and amphibians and so the default water ingestion rate of 0.0001 L/day has been adopted (USEPA 1999). Water ingestion rate was calculated using the following formulas:

Birds 𝐵𝐵𝐼𝐼 = 0.059 ∗ 𝐵𝐵𝐼𝐼0.67

Mammals 𝐵𝐵𝐼𝐼 = 0.099 ∗ 𝐵𝐵𝐼𝐼0.90

Where:

WI = drinking water ingestion (L/d)

Wt = body weight of receptor species (kg)

Calculations, inputs and results are shown in Appendix N.

4.3.2. Wildlife Exposure Parameters Tables 4-5 and 4-6 provide a summary of the wildlife exposure assumptions that have been adopted in the quantitative food web modelling. These assumptions include body weight, home range, calculated food and water ingestion rates based on equations presented in section 4.3.1, and diet composition. Many ROCs chosen may be either migratory or nomadic or have home ranges that are much larger than the IA, the screening assessment assumes that they spend 100% of the time in the investigation area. Dietary compositions, including uptake of soil or sediment, was based on dietary sources noted in Tables 3-8 and 3-9. Expert judgement was also applied in regards to the diet compositions adopted from advice provided by field biologists.



60

Table 4-5: Summary of adopted exposure parameters – aquatic ROCs Receptor of Concern (Common Name)

Assumed Body Weight (kg wet

weight) (b)

Food Ingestion Rate (kg wet weight/kg

body weight/day) (c)

Water Ingestion Rate (L/kg body weight/day) (d)

Diet Composition (%) (a)

A B C D E F G Invertivorous and Omnivorous Aquatic Birds

Eastern Great Egret 0.7 0.27 0.07 - 15 15 60 4 4 2

White-faced Heron 0.5 0.30 0.07 - 33 33 33 - - 1

Comb-crested Jacana 0.085 males 0.52 0.13 75 22 - - - - 3

Pacific Black Duck 1.025 (female) 0.24 0.059 85 12 - - - - 3

Radjah Shelduck 1.0 0.24 0.06 50 47 - - - - 3

Piscivorous Birds

Australasian Darter 2.6 0.18 0.04 - - - 100 - - -

Little Black Cormorant 0.52 0.30 0.07 - 25 - 75 - - -

Little Pied Cormorant 0.49 0.30 0.07 33 33 - 33 - - 1

Black-Necked Stork 4 0.16 0.04 - 33 33 33 - - 1

Piscivorous Mammals

Large-footed Myotis 0.015 0.43 0.15 - 50 - 50 - - -

Piscivorous Aquatic Reptiles

Gray’s Water Snake 0.08205 0.02 0.0001 - 40 - 50 - - 10

Keelback Snake 0.250 0.023 0.0001 - - 70 15 15 - -

Northern Snake-necked Turtle 3.3 (males) 0.019 0.0001 24 24 24 24 - - 4

Omnivorous Aquatic Reptiles

Northern Snapping Turtle 7.5 0.018 0.0001 80 5 - 5 5 - 5

Northern Yellow-faced Turtle 3.3 (males) 0.019 0.0001 55 39 - - - - 6 a. The diet composition of each ROC has been determined based on the information and references presented in Appendix M. For the purposes of the food web modelling, sediment ingestion

rates have been assumed to make up a portion of the total dietary intake (as per USEPA, 1999 methodology) A = Plants (roots, shoots, seeds, nectar) B = Invertebrates (crustaceans, molluscs, etc.) C = Reptiles and amphibians D = Fish E = Mammals F = Birds G = Sediment

b. The assumed body weight for each are as outlined in Appendix M. c. Calculated based on equations presented in Section 4.3.1. The maximum reported body weight was used in the numerator and the minimum reported body weight was used in the denominator

to provide the most conservative estimates for food ingestion rates. d. Calculated based on equations presented in Section 4.3.1.



61

Table 4-6: Summary of adopted exposure parameters – terrestrial ROCs Receptor of Concern (Common Name)

Assumed Body Weight

(kg wet weight) (b)

Food Ingestion Rate (kg wet

weight/kg body weight/day) (c)



A B C D E F G


Rose-crowned fruit dove 0.105 0.49 0.12 98 - - - - - 2

Australasian Figbird 0.128 0.46 0.12 74 24 - - - - 2

Magpie Goose (e) 2.0

0.19 0.05 90 - - - - - 10

Bar-shouldered Dove(e) 0.13 0.45 0.12 90 - - - - - 10

Plumed Whistling-Duck(e) 1.0 0.24 0.06 90 - - - - - 10


Pale Field Rat (f) 0.08 0.29 0.13 98 - - - - - 2

Agile Wallaby 9.0

0.10 0.11 90 - - - - - 10

Black Flying Fox 0.667 0.18 0.10 100 - - - - - -

Black-footed Tree-rat 0.83 0.17 0.10 98 - - - - - 2

Grassland Melomys 0.055 0.32 0.13 98 - - - - - 2

Invertivorous and Omnivorous Terrestrial Birds

Masked Lapwing 0.37 0.33 0.08 - 70 - 20 - - 10

Straw-necked Ibis 1.1 0.23 0.06 - 70 - 20 - - 10

Black-faced Wood Swallow 0.035 0.68 0.18 20 80 - - - - -

Invertivorous and Omnivorous Mammals

Short-beaked Echidna 2.5 0.13 0.09 - 90 - - - - 10

Common Rock-Rat 0.025 0.38 0.14 45 45 - - - - 10

Northern Brown Bandicoot 2.1 0.14 0.09 35 35 15 - 5 - 10

Large Bent-winged Bat 0.008 0.5 0.16 - 100 - - - - -

Orange Leaf-nosed Bat 0.0065 0.52 0.16 - 100 - - - - -



62

Receptor of Concern (Common Name)

Assumed Body Weight

(kg wet weight) (b)

Food Ingestion Rate (kg wet

weight/kg body weight/day) (c)



A B C D E F G

Invertivorous and Omnivorous Reptiles

Bynoe’s Gecko 0.118 0.024 0.0001 - 98 - - - - 2

Gilbert’s Dragon 0.9 0.021 0.0001 - 98 - - - - 2

Striped Rainbow Skink 0.3 0.023 0.0001 - 89 99 - - - 2


Black-shouldered Kite 0.291 0.35 0.089 - 10 - - 90 - -

Wedge-tailed Eagle 4 0.16 0.037 - - 10 - 80 10 -

Whistling Kite 0.77 0.26 0.064 - 25 - 25 25 25 -

Black Kite 0.65 0.28 0.068 - 33 33 - 34 - -


Northern Quoll 0.3 0.22 0.112 20 20 20 - 20 15 5

Dingo 12 0.09 0.077 - 10 9 - 75 5 1


Floodplain Monitor 1.35 0.029 0.0001 - 75 25 - - - -

Olive Python 10 0.018 0.0001 - - 33 - 33 34 -

Brown Tree Snake 2.3 0.02 0.0001 - - 50 - - 50 - a. The diet composition of each ROC has been determined based on the information and references presented in Appendix M. For the purposes of the food web modelling, sediment

ingestion rates have been assumed to make up a portion of the total dietary intake (as per USEPA, 1999 methodology) A = Plants (roots, shoots, seeds, nectar) B = Invertebrates (earthworms, insects, etc.) C = Reptiles and amphibians D = Fish E = Mammals F = Birds G = Soil

b. The assumed body weight for each are as outlined in Appendix M. c. Calculated based on equations presented in Section 4.3.1. d. Calculated based on equations presented in Section 4.3. e. Diet composition for the magpie goose, bar-shouldered dove, and plumed whistling duck were assumed to be like the Canadian goose and the diet composition including the percentage

of incidental soil uptake was used as referenced in Beyer et al 1994 (was cited as Beyer (in press) in the 1993 US EPA Wildlife Exposure Factors Handbook). f. Diet composition for the pale field rat were assumed to be like the white-footed mouse and the diet composition including the percentage of incidental soil uptake was used as referenced

in Beyer et al 1994 (was cited as Beyer (in press) in the 1993 US EPA Wildlife Exposure Factors Handbook).



63

5. Toxicity Assessment This section outlines the process used to evaluate abiotic and biotic media collected at the site to determine if PFAS compounds are potentially adversely impacting ecological receptors. This assessment was completed in three steps and the following sections outline the assessment values used in each step:

• Step 1 - Section 5.1 Generic guideline values were used in screening abiotic environmental media including surface soil, and surface water to indicate pathways that warranted further consideration associated with direct exposures;

• Step 2 - Section 5.2 Specific toxicity benchmarks (typically study toxicity NOAELs) were used to assess the reported environmental abiotic media against the exposure concentrations reported in studies to not result in adverse effects by specific lower trophic level organisms. The purpose of this step is to put the exposure concentrations in context with toxicity study levels; and

• Step 3 - Section 5.3 Specific study toxicity values (NOAELs) associated with intake of PFAS were used to put calculated intake for ROCs into context with reported toxicity study levels. Guideline toxicity reference values were also applied, where relevant, to indicate the point of negligible intake.

5.1. Step 1 – Generic Guideline Values Ecological Guideline Values as defined in the PFAS NEMP (2018) guidelines are available for PFAS compounds. The NEMP guideline values adopted are compared to abiotic sampling data to provide an initial indication of exposure areas that may pose a potential risk and that warrant further evaluation.

Guideline values have been selected for surface water and soils based on the identified receptors.

Guideline values are necessarily conservative and incorporate uncertainty factors over the raw toxicity values to ensure that a range of situations and species are protected. Exceedance of a guideline value does not necessarily indicate an elevated risk of adverse effects, but rather that the exposure should not be excluded from monitoring and further setting specific consideration.

5.1.1. Surface Water

Contaminants can present risks to water column communities and upper-trophic level receptors through direct contact and incidental ingestion of surface waters. Potential toxicity endpoints include adverse impact on survival, growth, and reproduction. In this instance, aquatic ecosystem protection values will be applied in the step 1 evaluation to conservatively consider water column community exposure. Guideline values for selected PFAS compounds, derived to be protective of aquatic ecosystems was sourced from:

• PFAS National Environmental Management Plan (HEPA, 2018)

The adopted Surface Water guideline values are summarised in Table 5-1.



64

Table 5-1: Surface water (freshwater and marine) guideline values Contaminant Aquatic Ecosystem Protection1

Freshwater Marine

Perfluorooctane sulfonate (PFOS) (µg/L) 0.13 0.13

Perfluorohexane sulfonate (PFHxS) (µg/L)

Perfluorooctanoic acid (PFOA) (µg/L) 220 220

1 PFAS NEMP (2018) Investigation levels for freshwater and marine exposure scenario at the 95% species protection – slightly to moderately disturbed systems.

Table 2 in Appendix G summarises the results of the comparison of each surface water sample result to the screening values in Table 5-1. There were no concentrations of PFOA measured that exceeded the 220 µg/L PFOA aquatic ecosystem protection concentration (HEPA, 2018) in freshwater or marine. In relation to surface water samples, for both freshwater and marine, PFOS was detected in 147 of the 195 samples collected. In relation to the 147 samples with detectable concentrations of PFOS, 89 samples (60% of samples), had concentrations of PFOS exceeding the 0.13 µg/L assessment criteria.

5.1.2. Soil

Contaminants can present risks to plants, soil invertebrates and upper-trophic level receptors through direct contact and incidental ingestion of soil. In this instance, terrestrial soil guideline values will be applied in the generic direct toxicity screening evaluation to conservatively consider soil invertebrate and plant exposure. Indirect exposure has been assessed through review of surface water and biota data that may have been influenced by soil contamination. Guideline values have been sourced from PFAS National Environmental Management Plan (HEPA, 2018). The PFAS NEMP (HEPA, 2018) includes interim soil ecological guideline values for both direct exposure and indirect exposure, as summarised in Table 5-2. For the purpose of the direct toxicity evaluation for this ERA the Soil Guideline Values for direct exposure have been adopted. The guideline values based on indirect exposures (i.e. based on soil ingestion by a secondary consumer and discharge to a surface water body) have not been adopted for the direct toxicity screening in this ERA. It is recognised that a significant proportion of the soil results for PFOS exceed the guideline values based on indirect exposure. As indirect exposures related to PFOS in soil have been specifically evaluated through other aspects of this ERA, including through food web modelling and evaluation of exposures to PFOS in surface waters which have been impacted by the mobilisation of PFOS in soil.

Table 1 in Appendix G summarises the results of the comparison of each surface soil (<0.5 m bgl) sample result to the screening values in Table 5-2. There were no concentrations of PFOA measured that exceeded the 10 mg/kg PFOA soil assessment criteria for the protection of terrestrial ecosystems by direct exposure (HEPA, 2018). With respect to measured surface soil (<0.5 m bgl), PFOS was detected in 469 of 676 samples collected. Of those 469 samples with detected concentrations of PFOS, 85 or 18% exceed the 1 mg/kg guideline value for ecological direct exposure in public open spaces.



65

Table 5-2: Soil guideline values.

Shading indicates general soil guideline value adopted for direct toxicity screening; ND = Not determined

1 - Does not take into account water transport. 2 - Interim soil screening values for ecological direct exposure were adopted from human health screening values for public

open space.

3 – Interim soil screening value for ecological indirect exposure adopted from 2017 Canadian Federal Environmental Quality Guidelines for Residential and Parkland, based on soil ingestion by a secondary consumer.

4 – Interim soil screening value for ecological indirect exposure adopted from 2017 Canadian Federal Environmental Quality Guidelines for Commercial and Industrial – Coarse Soil, based on concentration in soil that is expected to protect against potential impacts on freshwater life from PFOS originating in soil that may enter the groundwater and subsequently discharge to a surface water body.

5.1.3. Sediment

Contaminants in sediment may represent a potential on-going source of contamination to surface water, and present direct exposure to benthic invertebrates, plants and indirect exposure to other aquatic biota (i.e., amphibians, fish, etc).

For freshwater and marine sediments screening values for PFOS and PFOA have not been established.

5.2. Step 2 – Specific Direct Toxicity Comparison – Lower Trophic Level Organisms

For areas where generic Tier 1 guideline values are exceeded, additional methods of assessing likelihood of direct toxicity to specific groups of organisms have been used such as the use of direct study toxic effects results as benchmarks. For example, the risk of direct toxicity to fish has been assessed based on comparison of concentrations of PFOS and PFOA in water to reported lower bound sub-lethal toxic effect concentrations to specific fish species (i.e. No Observed Effect Concentration (NOEC) and Lowest Observed Effect Concentration (LOEC) values) determined from controlled laboratory studies. This is broadly consistent with the approach in the US EPA Ecological Risk Assessment Guidance for Superfund (US EPA, 1997) and US EPA Ecological Risk Assessment Process in the Office Pesticides Programs (US EPA, 2004).

The main objective of direct toxicity risk characterisation is to determine if lower order organisms are at risk of material adverse effects. As many of the lower order organisms assessed may be prey items for higher order organisms, the evaluation of direct toxicity that would lead to reduced reproduction, development delay or death, in part also evaluates potential indirect effects to higher order organisms through the potential loss of prey items. As part of this risk review, the exposure concentrations (soil and water) were compared against direct toxicity benchmarks from the body of available toxicity testing studies.

Contaminant Terrestrial Soil Ecological Guideline Values

Ecological Direct Exposure (1)(2)

Ecological Indirect Exposure – Soil ingestion by a secondary consumer

(3)

Ecological Indirect Exposure – Discharge to

surface water body (4)

PFOS (mg/kg)

1 0.01 0.14

PFOA (mg/kg)

10 ND ND



66

The direct toxicity benchmarks adopted have been selected following a review and evaluation of available direct toxicity data sourced from the literature, as per the approach outlined in Section 1.5.1. The direct toxicity data considered is detailed in Appendix L. The compiled literature-based toxicity data has been sorted by taxonomic group (i.e., terrestrial invertebrates, aquatic freshwater plants, birds, etc.). Every attempt was made to obtain the primary data that referenced the endpoint. The primary data were assessed using a modified Hobb’s scale (Hobbs et al., 2005) as presented in Batley et al. (2014).This method is based on the method used in the AQUIRE (ECOTOX) (USEPA, 1994) database. Only ‘high’ (scoring 80-100) or ‘acceptable’ quality data sets (scoring 51-79) are deemed suitable for use (Batley et al., 2014). Scores based on the Hobbs scale are presented with each reference in Appendix L. The direct toxicity benchmarks adopted are provided in Table 5-3.

The direct toxicity risk review applied here relies on chronic effects endpoints that are either direct measures of impairment of survival, growth, or fecundity or other endpoints for which there is a scientifically robust, peer reviewed relationship that can quantify the impact of the measured effect endpoint on the assessment endpoints of survival, growth, and fecundity (USEPA 2004). Thus, certain types of individual organism effects, such as physiological measures or organism behaviour changes, have generally not included in the effects analysis because the linkage between these individual effects and population-level effects is uncertain and not quantitative given the present state of knowledge. USEPA (1997) guidance specifies that a direct toxicity benchmark value should be “equivalent to a documented or best conservatively estimated chronic No Observed Adverse Effect Level (NOAEL).” In some cases, when only the lowest observed effect level or concentration (LOEL/LOEC) was documented, an uncertainty factor of 10 was applied to conservatively estimate the no observed effect level or concentration (NOEL/NOEC). This was the only uncertainty factor applied to the compiled toxicity values for the purpose of the direct toxicity screening evaluation. The direct toxicity benchmark value adopted was considered to be a chronic NOAEL (or NOEC) for the most sensitive species tested in relation to an ecologically relevant endpoint.

Direct effects were evaluated by comparing the maximum media concentration for a given area with compiled chronic toxicity data for species that are considered surrogates for site-specific species. Toxicity data is generally lacking for most site-specific organisms; therefore, effects data were obtained from surrogate species that result in the most sensitive endpoint (i.e., lowest NOEC) for that species group. Although there may be data for species more closely related taxonomically and ecologically, the more sensitive endpoint was used as the surrogate as a conservative measure.

In relation to the following types of organisms, we have not adopted specific direct toxicity benchmarks due to insufficient toxicological data, of sufficient quality, being available:

• Sediment: Freshwater plants.

• Sediment: Freshwater benthic invertebrates.

• Sediment: Marine / Estuarine plants.

• Sediment: Marine / Estuarine benthic invertebrates.

• Surface water: Amphibians.

• Surface water: Marine / Estuarine plants.

In these instances concentrations in media were compared to generic guideline values that address the broader environment, where available, for the purposes of the direct toxicity evaluation.



67

Table 5-3: Summary of direct toxicity benchmarks

Exposure Scenario

Analyte Direct Toxicity Benchmark

Reference/Justification

Soil

Terrestrial Organisms (general – in the absence of available soil screening benchmarks, this value has been adopted to assess the potential for exposures to plants and animals)

PFOS 1 mg/kg PFAS National Environmental Management Plan (January 2018), as an interim ecological guideline value based on the human health screening value for public open space (HEPA, 2018)

PFOA 10 mg/kg PFAS National Environmental Management Plan (January 2018)

As adopted from the 2017 Canadian Federal Environmental Quality Guidelines for Residential and Parkland (soil ingestion by a secondary consumer) and Commercial and Industrial – Coarse Soil (concentration in soil that is expected to protect against potential impacts on freshwater life from PFOS originating in soil that may enter the groundwater and subsequently be discharged to a surface water body.

Terrestrial Plants PFOS 4.6 mg/kg Based on lettuce No Observed Effect Concentration (NOEC) for growth endpoints (shoot height and weight) (Environment Agency, 2004).

PFOA 1.0 mg/kg Based on wheat No Observed Effect Concentration (NOEC) for growth endpoints (Zhao et al., 2014).

Terrestrial Invertebrates

PFOS 10 mg/kg Based on earthworm NOEC for reproductive endpoints (Norwegian Pollution Control Authority, 2008).

PFOA 10 mg/kg Based on earthworm NOEC for reproductive endpoints (He et al., 2016).

Surface Water

Freshwater/

Estuarine/

Marine

PFOS 0.00013 mg/L PFAS National Environmental Management Plan (January 2018),95% species protection justified based on urban environment.

PFOA 0.22 mg/L PFAS National Environmental Management Plan (January 2018)

Freshwater - plants

PFOS 2.9 mg/L Based on the 42-day NOEC for Myriophyllum sibiricum (Hanson et al., 2005).

PFOA 12.5 mg/L Based on the 96-hour NOEC to Pseudokirchneriella subcapitata for growth (OECD 2008).

Freshwater – invertebrates

PFOS 0.01 mg/L Based on the 120-day NOEC for Enallagma cyathigerum survival (Bots et al., 2010)

PFOA 0.01 mg/L Based on the PFOS 120-day NOEC for Enallagma cyathigerum survival (Bots et al., 2010) due to the lowest PFOA NOEC being orders of magnitude higher (i.e. 20 mg/L for Daphnia magna reproduction (OECD 2008)).

Freshwater – fish PFOS 0.0003 mg/L Based on 90-day LOEC for egg development of Danio rerio, extrapolated to a chronic NOEC of 0.3 µg/L (referenced in draft ANZECC Water Quality Default Guideline Values for PFOS (DoEE 2016)

PFOA NA Lowest PFOA NOEC was 40 mg/L for Oncorhynchus mykiss survival (OECD 2008).

Marine/estuarine - invertebrates

PFOS 0.0001 mg/L Based on M. galloprovincialis 96-hour LOEC for developmental endpoints (Fabbri et al. 2014).

PFOA 0.0001 mg/L

Marine/estuarine fish

PFOS 0.015 mg/L Based on Psetta maxima 144-hour NOEC for developmental and survival endpoints (Mhadhbi et al. 2012).

PFOA 1.5 mg/L



68

5.3. Toxicity Values – Upper Trophic Level Organisms To evaluate the potential toxic effects to upper trophic level organisms we have identified toxicity values, based on dose (i.e. mass of chemical per unit of body weight) for comparison to estimates of the dietary intake of PFAS for ROC determined from the food web modelling. We conducted a literature search to identify toxicity values for PFAS expressed as milligrams per kilogram body weight per day ([mg/kg-body weight/day]) for avian, mammalian, reptilian and amphibian species. This toxicity has been compiled into a database as present in Appendix L. Every attempt was made to obtain the primary data that the endpoint was referenced. The primary data were assessed using a modified Hobb’s scale (Hobbs et al., 2005) as presented in Batley et al. (2014).This method is based on the method used in the AQUIRE (ECOTOX) (USEPA, 1994) database. Scores based on the Hobbs scale are presented with each reference in Appendix L. Based on the database of toxicity data compiled we selected toxicity values for upper trophic level receptors of concern, as described further below.

In selecting the toxicity values to be adopted for the upper trophic level risk assessment we considered the guidance in the US EPA Ecological Risk Assessment Guidance for Superfund (US EPA, 1997). The USEPA guidelines (US EPA, 1997) recommend the following in selecting ecotoxicity values for preliminary ERAs:

• Should represent a no-observed-adverse-effect-level (NOAEL) for long-term (chronic) exposures.

• Preference should also be given to ecological effects that can impact populations (or higher levels of biological organisation), such as adverse effects on development, reproduction, and survivorship.

• Preference for dietary dose-response studies over other less relevant methods of dosage (i.e., egg injection, dermal painting or in vitro studies)

• If exposure at the site is by ingestion, dietary studies are preferred to gavage studies, which are subsequently preferred to non-ingestion routes of exposure

• Preference for toxicity data from a dose-response study based on a range of dietary doses administered to a test species during a sensitive life stage (i.e., embryo, juvenile, or reproductive adult) over a chronic exposure duration.

• Preference for sensitive sub-lethal end points including reproduction, growth and development.

The above recommended preference from the US EPA guidelines (US EPA, 1997) have been generally followed in selecting the appropriate toxicity values to adopt for this ERA. It is noted that community-level effects are also important, however toxicity data on community level endpoints are limited and extrapolation from one community to another can be difficult (US EPA, 1997). As such community level toxicity data hasn’t specifically been considered as part of this ERA.



69

Depending on the volume and suitability of the available data, toxicity values for ecological receptors may be adopted for use based on one of two different approaches including: • For chemicals that have been well studied, there may be numerous high-quality dose-response

studies available including those that used multiple test species within a single taxonomic class and cover a wide range of doses, exposure durations and sensitive endpoints. In these cases, toxicity values may be adopted by selecting no observed adverse effect level (NOAEL) and lowest observed adverse effect level (LOAEL) values for individual receptor species or groups of interest at a project site using several critical studies (e.g. for specific species types within a class of receptors); or

• For chemicals that have been less well studied or are emerging there may be far fewer qualified studies available and when available these studies are typically limited to a few conventional test species and exposure conditions. As such generic NOAELs and LOAELs that are protective of multiple species within a class (e.g. birds, mammals) may be adopted by selecting the highest NOAEL or lowest LOAELs or by estimating the geometric mean of the NOAEL and LOAEL. The test species used in the studies may not be directly comparable to the trophic levels and body weights of the receptor groups to be examined at the site. In these cases, the reported endpoint NOAELs and LOAELs may be modified by Uncertainty Factors (UFs), as appropriate. The most sensitive of the few NOAELs and LOAELs is then selected as the generic NOAEL and LOAEL to be protective of the entire class (e.g., birds, mammals).

For the PFAS compounds considered in this ERA, reliable dose-response studies were available for relatively few different species of birds, mammals, reptiles and amphibians. Therefore, as noted in the second approach described above, a single NOAEL and LOAEL value are proposed to represent all birds, regardless of trophic level; similarly, a single set of NOAEL and LOAEL values are proposed for all mammals, regardless of trophic level. In the absence of specific toxicity data, avian toxicity data was used as a surrogate for terrestrial-phase reptiles and amphibians based on research conducted by Kannan et al. (2005). The rationale for this is described further below.

Like birds and mammals, reptiles are potentially exposed to contaminants through their diet (prey). Reptiles are present in the Darwin IA and in surrounding ecological environments and were sampled as part of the vertebrate tissue sampling. Reptilian ecological risks are generally not addressed through a direct, quantitative evaluation because dose-response data for reptile species are limited, not only for emergent chemicals, including PFAS, but for conventional toxins as well. Generally, reptiles are thought to have lower dietary exposure than birds and mammals because they have slower metabolic rates and are expected to ingest less food than birds and mammals (Weir et al, 2016). The toxic mode of action of PFAS in reptiles is not well understood and represents an uncertainty in the ecological risk assessment. As noted, a limited number of toxicity studies for reptiles were identified. Of the two identified studies, researchers determined bioaccumulation factors based on PFOS in blood serum of freshwater turtles and a simultaneously sampled 2 L volume of water (National Institute for Public Health and the Environment [RIVM], 2010). Researchers also sampled multiple trophic levels in the food web of the North American Great Lakes including snapping turtles as well as Chinook salmon, carp, and birds/mammals that prey on turtles (mink and bald eagles) (Kannan et al., 2005). RIVM (2010) indicates that reptiles are being exposed and may be accumulating PFOS but direct toxicity values could not be extrapolated. Whereas, the Kannan (2005) study indicates that concentrations of PFOS in the snapping turtles (and Chinook salmon and carp) are, up to an order of magnitude less than those in trophic levels above them that prey on turtles (i.e, mink, bald eagles). Kannan (2005) also noted a gender difference in the concentrations of PFOS in snapping turtles suggesting oviparous transfer of PFOS through egg laying, like that observed for birds and fish (Kannan et al., 2005). Therefore, it is assumed for this risk assessment that toxicity reference values for top-level predatory birds may also be used to evaluate the potential risk to reptiles. The avian NOAEL and LOAEL have been used to evaluate PFOA/PFOS potential effects on reptilian receptors.



70

The adopted toxicity values are summarised in Table 5-4. In relation to avian receptors, relevant toxicity data for PFOS was identified for two different avian species being Northern Bobwhite Quail (Colinus virginianus) and the Mallard (Anas platyrhynochos). The Northern Bobwhite Quail was determined to be the most sensitive of these two avian species with a chronic LOAEL of 0.77 mg/kg/bw-day (PFOS) for sub-lethal reproductive endpoints being reported. An uncertainty factor of 10 was applied to this LOAEL to extrapolate to a NOAEL. In the absence of reliable PFOA specific toxicity data the PFOS toxicity values have been adopted as a surrogate for the PFOA toxicity evaluation.

In relation to mammalian receptors, relevant toxicity data was identified for four different species being cynomolgus monkeys (Macara fasicularis), rabbits (Oryctolagus cuniculus), mice and rats. For PFOS the lowest toxicity values were determined from a study of rats and rabbits with a LOAEL of 0.4 mg/kg/bw-day and a NOAEL of 0.1 mg/kg/bw-day for chronic reproductive endpoints reported. For PFOA the lowest toxicity values were determined from a study of mice and rats with a LOAEL of 7.6 mg/kg/bw-day and a NOAEL of 6.2 mg/kg/bw-day for chronic reproductive endpoints reported. For the purpose of additional context in this ERA, estimated intake doses have also been compared to the Tolerable Daily Intake (TDIs) for PFOS used for the derivation of the Canadian Federal Environmental Quality Guidelines for Wildlife Diet (ECCC, 2018). These TDIs are considered to be conservative screening guideline values. The TDIs are based on the same toxicity dataset as the LOAELs and NOAELs adopted for this ERA, however for the purpose of deriving the Canadian screening level diet values an additional uncertainty factor of 10 was applied to the extrapolated NOAEL to provide an additional level of certainty that the wildlife diet guideline values were protective of all avian species.

Table 5-4: Summary of adopted avian and mammalian toxicity values Analyte Toxicity value (mg/kg BW/day) Toxicity Endpoint Reference/Justification

TDI (Screening

value)

NOAEL LOAEL

Avian (also adopted to assess risks to reptiles)

PFOS 0.0077 0.077 0.77 Reproductive endpoints (e.g. chick survival)

LOAEL for reproductive endpoints (e.g., chick survival) for the Bobwhite Quail (Newsted et al., 2005). An uncertainty factor of 10 was used to convert a LOAEL to a NOAEL.

PFOA 0.0077 0.077 0.77 PFOS used as surrogate in absence of chemical-specific toxicity values

Mammalian

PFOS 0.0011 0.1 0.4 NOAEL and LOAEL for reproductive endpoints (e.g. decreased litter size, birth weight and pup survival, and developmental abnormalities) based on singular and multi-generational studies

Based on results of four studies on rats and rabbits (Environment Canada, 2006; Stahl et al., 2011; RIVM, 2010; Dietz et al., 2015). No uncertainty factors applied to the NOAEL or LOAEL.

PFOA 0.0011 6.2 7.6 Reproductive endpoints (e.g. reduced pup body weight, behavioural effects [reduced motor coordination], vascular mineralisation in testes, tubular hyperplasia in ovaries)

The toxicity values are the calculated geometric mean of LOAEL and NOAEL values from eight mammalian studies on mice and rats (ASTDR, 2015). No uncertainty factors applied to NOAEL and LOAEL values.



71

6. Risk Characterisation Risk characterisation is the process used to estimate the potential for adverse ecological impacts based on information obtained in the exposure (Section 4) and effects (Section 5) sections. Direct and indirect toxicity is evaluated. Direct toxicity is evaluated by comparing the maximum detected and average concentrations in a certain media to the ecological guideline value for generic environments, and against study toxicity values for certain biota including terrestrial plants, terrestrial invertebrates, freshwater plants and invertebrates, freshwater fish, and estuarine invertebrates, and estuarine fish.

6.1. Risks Associated with Direct Toxicity Risks associated with direct toxicity are typically evaluated with respect to lower order organisms (e.g., plants, terrestrial and aquatic invertebrates, fish). This level of organism tends to be the most highly exposed to PFAS that may be present in the abiotic media including soil, sediment and surface water.

Maximum detected concentrations were used in the direct toxicity evaluation as this was the most conservative. By using the maximum detected concentration, any effect on lower order organisms will be highlighted. When there is no direct toxicity noted based on the maximum detected concentration, the conservativeness of the exercise allows that that media in that EEA to be eliminated in terms of potential adverse effects on that class of organisms.

PFOS concentrations in soil and surface water are presented in Figures 6, 7 and 8 of Appendix A relative to the direct toxicity guideline values.

6.1.1. Terrestrial Plants and Invertebrates Tables 6-1 and 6-2 summarises the results of the reported PFOS and PFOA concentrations in shallow surface soil (0.0 – 0.5 m bgl) in the four terrestrial ecological exposure areas to plants (Table 6-1) and invertebrates (Table 6-2). Only shallow soils, 0.0 – 0.5 m bgl, have been evaluated because this is considered the most biologically active zone and where ecological receptors would likely be exposed.

The results of the evaluation of direct toxicity of PFOS to terrestrial plants indicated that there is potential for direct toxicity in the form of reduced growth in localised areas in each of the on Base EEA, except for the Zone A EEA (Table 6-1). The maximum soil concentration in the Zone D EEA indicated a ratio of the maximum soil concentration to the direct toxicity benchmark of 15 or the maximum soil concentration was 15.3 times above the plant direct toxicity benchmark. The fraction of reported PFOS concentrations that was greater than the plant direct toxicity benchmark was less than 3.0% across the entire investigation area (2.8%), with the Zone B EEA with the highest, 4.4%, and the other EEAs having less than 3.0%. The average concentration calculated for each of the EEA was less than plant direct toxicity benchmark, thus the direct toxicity may be limited in spatial extent across the EEAs. Results above the terrestrial plant benchmark are shown in orange and red in Figures 6a to 6g in Appendix A.

The results of the evaluation of direct toxicity of PFOA to terrestrial plants indicated that direct toxicity is unlikely as the maximum measured soil concentration in any EEA was less than the terrestrial plant direct toxicity benchmark (Table 6-1).



72

The results of the evaluation of direct toxicity of PFOS to terrestrial invertebrates indicated that there is potential for direct toxicity in the form of reduced reproduction in areas within Zone B and Zone D (Table 6-2). The maximum soil concentration in the Zone D was 7.1 times higher than the toxicity benchmark. The proportion of reported PFOS concentrations greater than the invertebrate direct toxicity benchmark was less than 2.0% across the entire IA (1.6%), with Zone B with the highest, 4.4%, and the other EEAs having less than or equal to 2.0%. The average concentration calculated for each of the EEAs was less than the invertebrate direct toxicity benchmark, thus the direct toxicity is likely to be limited to isolated areas around sources. Results above the terrestrial invertebrate benchmark are shown in red in Figures 6a to 6g in Appendix A.

The results of the evaluation of direct toxicity of PFOA to terrestrial invertebrates indicated that direct toxicity is unlikely as the maximum measured soil concentration, as well as all soil concentrations measured, in each EEA was less than the terrestrial invertebrate direct toxicity benchmark (Table 6-2).



73

Table 6-1: Summary of surface soil concentrations and potential for direct toxicity to terrestrial plants Assessment Area

Minimum Detected Concentration (mg/kg)

Maximum Detected Concentration (mg/kg)

Average Detected Concentration (mg/kg)

Detects (Total Number of Samples)

Direct Toxicity Benchmark (mg/kg)

Ratio of Max Detected / Plant Benchmark – ratios >1 indicate potential for toxicity

Number of Samples Exceeding Plant Benchmark (Frequency of Exceedance – percentage of samples greater than benchmark)

PFOS

All 0.0005 70.59 1.2 539 (824) General – 1; Plants - 4.6

15.3 23 of 824 (2.8%)

Zone A 0.0096 0.035 0.02 13 (29) 0.007 0 of 29 (0.0%)

Zone B 0.0014 39.0 1.9 109 (137) 8.5 6 of 137 (4.4%)

Zone C 0.0023 9.1 0.9 179 (310) 2.0 7 of 310 (2.3%)

Zone D 0.0005 70.59 1.1 238 (348) 15.3 10 of 348 (2.9%)

PFOA

All 0.0005 0.520 0.03 128 (824) General – 10; Plants – 1.0

0.52 0 of 824 (0.0%)

Zone A ND (<0.005) ND (<0.005) ND (<0.005) 0 (29) 0.005 0 of 29 (0.0%)

Zone B 0.001 0.327 0.03 66 (137) 0.327 0 of 137 (0.0%)

Zone C 0.0005 0.52 0.05 26 (310) 0.52 0 of 310 (0.0%)

Zone D 0.0006 0.12 0.03 36 (348) 0.12 0 of 348 (0.0%)

Bolded values exceed the general soil screening value. Highlighted values indicate that the maximum detected soil concentration exceeded the plant benchmark.



74

Table 6-2: Summary of surface soil concentrations and potential for direct toxicity to terrestrial invertebrates Assessment Area

Minimum Detected Concentration (mg/kg)

Maximum Detected Concentration (mg/kg)

Average Detected Concentration (mg/kg)


Direct Toxicity Benchmark (mg/kg)

Ratio of Max Detected / Invertebrate Benchmark – ratios >1 indicate potential for toxicity

Number of Samples Exceeding Invertebrate Benchmark (Frequency of Exceedance – percentage of samples greater than benchmark)

PFOS

All 0.0005 70.59 1.2 539 (824) General – 1; Invertebrates

– 10

7.1 13 of 824 (1.6%)

Zone A 0.0096 0.035 0.02 13 (29) 0.0035 0 of 29 (0.0%)

Zone B 0.0014 39.0 1.9 109 (137) 3.9 6 of 137 (4.4%)

Zone C 0.0023 9.1 0.9 179 (310) 0.91 0 of 310 (0.0%)

Zone D 0.0005 70.59 1.1 238 (348) 7.1 7 of 348 (2.0%)

PFOA

All 0.0005 0.520 0.03 128 (824) General & Invertebrates

– 10

0.052 0 of 824 (0.0%)

Zone A ND (<0.005) ND (<0.005) ND (<0.005) 0 (29) 0.0005 0 of 29 (0.0%)

Zone B 0.001 0.3268 0.03 66 (137) 0.033 0 of 137 (0.0%)

Zone C 0.0005 0.520 0.05 26 (310) 0.052 0 of 310 (0.0%)

Zone D 0.0006 0.120 0.03 36 (348) 0.012 0 of 348 (0.0%) Bolded values exceed the general soil screening value. Highlighted values indicate that the maximum detected soil concentration exceeded the invertebrate s benchmark .



75

6.1.2. Freshwater Plants and Invertebrates Comparison of the maximum and average PFOS and PFOA concentrations in waterways across the investigation area against the generic guideline value indicated that there was potential for risk to freshwater aquatic ecosystems in Rapid Creek. The maximum and average PFOS concentrations both exceeded the guideline value which is designed to be protective of all aquatic species with a high level of confidence.

The results of the evaluation of direct toxicity of PFOS and PFOA in surface water to freshwater aquatic plants indicated that there is low potential for direct toxicity in any of the freshwater EEAs across the Investigation Area (Table 6-3). All surface water samples collected indicated PFOS and PFOA concentrations, either detected or level of reporting for non-detects, that were less than the freshwater aquatic plant surface water direct toxicity benchmark.

The results of the evaluation of direct toxicity of PFOS and PFOA in freshwater surface water to freshwater aquatic water column invertebrate communities indicated that there is low potential for direct toxicity in the freshwater EEAs across the investigation area (Table 6-4). All surface water samples collected indicated PFOS and PFOA concentrations, either detected or at the level of reporting for non-detects, that were less than the freshwater aquatic invertebrate surface water direct toxicity benchmark.

Due to the absence of reliable sediment based direct toxicity screening benchmarks it was not possible to evaluate the potential for direct toxic effects to freshwater plants and benthic invertebrates associated with exposure to PFOS and PFOA in sediment.

.



76

Table 6-3: Summary of freshwater surface water concentrations and potential for direct toxicity to freshwater aquatic plants Assessment Area

Minimum Detected Concentration (µg/L)

Maximum Detected Concentration (µg/L)

Average Detected Concentration (µg/L)


Direct Toxicity Benchmark (µg/L)

Ratio of Max Detected / Screening Benchmark – ratios >1 indicate potential for toxicity

Number of Samples Exceeding Screening Benchmark (Frequency of Exceedance – percentage of samples greater than benchmark)

PFOS

Rapid Creek 0.05 1.6 0.56 73 (73) General – 0.13; Plants – 2,900

0.0006 0 of 73 (0.0%)

PFOA

Rapid Creek 0.01 0.06 0.02 54 (73) General 220; Plants – 12, 500

0.000005 0 of 73 (0.0%)

Bolded values exceed the general soil screening value. Highlighted values indicate that the maximum detected soil concentration exceeded the aquatic plant benchmark



77

Table 6-4: Summary of freshwater surface water concentrations and potential for direct toxicity to freshwater aquatic water column invertebrate communities Assessment Area Minimum

Detected Concentration (µg/L)





Ratio of Max Detected / Benchmark – ratios >1 indicate potential for toxicity

Number of Samples Exceeding Benchmark (Frequency of Exceedance – percentage of samples greater than benchmark)

PFOS

Rapid Creek 0.05 1.6 0.56 73 (73) General – 0.13; Invertebrates - 10

0.16 0 of 73 (0.0%)

PFOA

Rapid Creek 0.01 0.06 0.02 54 (73) General – 6.8; Invertebrates - 10

0.006 0 of 73 (0.0%)

Bolded values exceed the general surface water screening value. Highlighted values indicate that the maximum detected surface water concentration exceeded the invertebrate benchmark.



78

6.1.3. Freshwater Fish Concentrations of PFOS and PFOA reported in surface water of the freshwater portion of Rapid Creek were evaluated for potential impacts to freshwater fish (Table 6-5).

Maximum and average concentrations from surface water samples collected from the impacted section of freshwater Rapid Creek exceeded the nominated fish toxicity benchmark, indicating potential for adverse effects on development of some species.

Table 6-5: Summary of freshwater surface water concentrations and potential for direct toxicity to freshwater fish Assessment Area Minimum








PFOS

Rapid Creek 0.05 1.6 0.56 73 (73) General – 0.13; Fish 0.3

5.3 62 of 73 (85%)

PFOA

Rapid Creek 0.01 0.06 0.02 54 (73) General – 220; Fish NA

- -

Bolded values exceed the general surface water screening value. Highlighted values indicate that the maximum detected surface water concentration exceeded the freshwater fish benchmark.



79

6.1.4. Estuarine Plants and Invertebrates The maximum surface water concentrations in Rapid Creek, Ludmilla Creek and upper reaches of Sadgroves and Reichardt Creeks were above the generic direct toxicity Guideline value, which is protective of all aquatic flora and fauna (Table 6-6).

Comparison of maximum and average PFOS and PFOA concentrations in estuarine waters against plant direct toxicity benchmarks indicated negligible risk to estuarine plants.

The maximum concentrations in all waterways and the average concentrations calculated for Rapid Creek Estuary and Ludmilla Creek was greater than the water column invertebrate direct toxicity benchmark (Table 6-7). The potential for direct toxicity effects is most likely in Ludmilla Creek with 40% of the estuarine samples exceeding the direct toxicity benchmark. The results of the evaluation of direct toxicity of PFOA in estuarine surface water to estuarine water column invertebrates indicated that there is low potential for direct toxicity in all EEAs except Ludmilla Creek (Table 6-7). However, the average surface water concentration in Ludmilla Creek was below the direct toxicity benchmark.

Due to the absence of reliable sediment based direct toxicity screening benchmarks it was not possible to evaluate the potential for direct toxic effects to estuarine plants and benthic invertebrates associated with exposure to PFOS and PFOA in sediment.

6.1.5. Estuarine Fish

Concentrations of PFOS and PFOA reported in surface water of the estuarine portion of Rapid Creek, Ludmilla Creek, and Reichardt and Sadgroves creeks were evaluated for potential impacts to estuarine fish (Table 6-8). The maximum surface water concentration did not exceed the adopted estuarine fish direct toxicity benchmark in any assessment area. However, we note that the maximum PFOS concentration in Ludmilla Creek and Rapid Creek, and the average concentration in Ludmilla Creek did exceed the Freshwater fish direct toxicity benchmark. The risk of adverse effects to estuarine fish associated with exposure to PFOS or PFOA is negligible in Sadgroves and Reichardt creeks, but low risk of developmental effects may be present to some fish species in upper portions of Ludmilla Creek and Rapid Creek.



80

Table 6-6: Summary of estuarine surface water concentrations and potential for direct toxicity to estuarine aquatic plants Assessment Area

Minimum Detected Concentration (µg/L)




PFOS




PFOS

All – Estuarine 0.003 6.6 0.87 73 (109) General – 0.13; Plants

– 2,900

0.0023 0 of 109 (0.0%)

Rapid Creek Estuary

0.027 0.66 0.15 14 (17) 0.00023 0 of 17 (0.0%)

Ludmilla Creek Estuary

0.003 6.6 1.81 33 (42) 0.0023 0 of 42 (0.0%)

Reichardt and Sadgroves creek

0.01 0.25 0.076 26 (50) 0.00009 0 of 50 (0.0%)

PFOA

All – Estuarine Dry

0.002 0.19 0.053 29 (109) General – 220;

Plants – 12,500

0.00002 0 of 109 (0.0%)

Rapid Creek Estuary

0.01 0.02 0.013 3 (17) 0.000002 0 of 17 (0.0%)

Ludmilla Creek Estuary

0.002 0.19 0.072 20 (42) 0.00002 0 of 42 (0.0%)


0.01 0.02 0.12 6 (50) 0.000002 0 of 50 (0.0%)

Bolded values exceed the general surface water screening value. Highlighted values indicate that the maximum detected estuarine surface water concentration exceeded the plant benchmark.



81

Table 6-7: Summary of estuarine surface water concentrations and potential for direct toxicity to estuarine water column invertebrates Assessment Area Minimum





PFOS




PFOS

All – Estuarine 0.003 6.6 0.87 73 (109) General – 0.13; Invertebrates –

0.1

66 5 of 40 (12.5%)

Rapid Creek Estuary 0.027 0.66 0.15 14 (17) 6.6 1 of 10 (10%)

Ludmilla Creek Estuary 0.003 6.6 1.81 33 (42) 66 17 of 42 (40%)


0.01 0.25 0.076 26 (50) 2.5 4 of 30 (13.3%)

PFOA

All – Estuarine 0.002 0.19 0.053 9 (67) General – 220; Invertebrates –

0.1

1.9 0 of 40 (0.0%)

Rapid Creek Estuary 0.01 0.02 0.013 2 (17) 0.2 0 of 10 (0.0%)

Ludmilla Creek Estuary 0.002 0.19 0.072 20 (42) 1.9 7 of 42 (17%)


0.01 0.02 0.012 6 (50) 0.2 0 of 30 (0.0%)

Bolded values exceed the general surface water screening value. Highlighted values indicate that the maximum detected estuarine surface water concentration exceeded the estuarine water column invertebrate screening benchmark.



82

Table 6-8: Summary of estuarine surface water concentrations and potential for direct toxicity to estuarine fish Assessment Area Minimum





PFOS




PFOS

All – Estuarine 0.003 6.6 0.87 73 (109) General – 0.13;

Fish – 15 (Freshwater fish – 0.3)

0.44 0 of 109 (0.0%)


Ludmilla Creek Estuary 0.003 6.6 1.81 33 (42) 0.44 0 of 42 (0.0%)


0.01 0.25 0.076 26 (50) 0.017 0 of 50 (0.0%)

PFOA

All – Estuarine 0.002 0.19 0.053 9 (67) General – 220;

Fish – 1500

0.00013 0 of 67 (0.0%)


Ludmilla Creek Estuary 0.002 0.19 0.072 20 (42) 0.00019 0 of 42 (0.0%)


0.01 0.02 0.012 6 (50) 0.000013 0 of 50 (0.0%)

Bolded values exceed the general surface water screening value. Highlighted values indicate that the maximum detected estuarine surface water concentration exceeded the estuarine fish benchmark.



83

6.1.6. Potential Contribution of Other PFAS to Direct Toxicity

The potential direct toxicity risks associated with exposure to PFAS in abiotic media has been based on exposure to PFOS and PFOA. However, as discussed in Section 4.2, other PFAS compounds have been detected in abiotic media, which may contribute to toxic effects, therefore the potential direct toxicity risks could be underestimated for some scenarios by not considering the contribution of other PFAS compounds. As such this section provides a qualitative evaluation of the potential for other PFAS compounds to contribute to toxicity and alter the overall estimates of risk for different scenarios.

In relation to soil, other PFAS compounds were found to represent less than 15% of the total PFAS concentration. For the majority of the direct toxicity scenarios related to exposure to soil and potential risk to indicate based on the maximum PFAS concentrations alone, hence any contribution from other PFAS compounds will not change the risk outcome. The PFOS and PFOA concentrations in soil are below the direct toxicity benchmarks for terrestrial plants in Zone A and terrestrial invertebrates in Zone A and C. Contribution of the other PFAS compounds to the direct toxicity in these scenarios is considered unlikely to alter the risk outcome. This is as the other PFAS compounds represent less than 15% of total PFAS, and assuming that the other PFOS compounds could be as toxic as PFOS, the total PFAS concentration would need to increase by two orders of magnitude to exceed the direct toxicity benchmarks for plants and invertebrates.

Other PFAS compounds were found to generally represent up to 50% of the total PFAS concentration in freshwater and estuarine surface waters. In relation to freshwater the maximum PFOS concentration in Rapid Creek is one to three orders of magnitude below the specific direct toxicity benchmarks for freshwater plants, invertebrates and fish. As such any contribution from other PFAS compounds is considered unlikely to change the risk outcome, in the absence of evidence that other PFAS compounds have a greater proportional toxic effect than PFOS. Similarly, contribution from other PFAS compounds is unlikely to change the interpretation of risk in relation to estuarine waters and potential direct toxic effects to plants and fish as the maximum PFOS concentrations are at least an order of magnitude below the specific direct toxic benchmark in Rapid Creek, Reichardt Creek and Sadgroves Creek. There is some potential that other PFAS compounds could contribute to a change in the direct toxicity risk outcome for estuarine fish in Ludmilla Creek as there is only a two-fold difference between the maximum PFOS concentrations and the specific direct toxicity benchmark.

6.2. Risks to Higher Trophic Levels (Food Web Modelling) The direct toxicity evaluation results presented in Section 6.1 and Section 6.2 indicate that there is limited evidence that reported PFOS/PFOA concentrations in the various environmental media are likely to be causing direct toxic effects, with some exceptions for marine/estuarine invertebrates and potentially fish in Ludmilla Creek, and terrestrial plants in key hotspots on-Base. However, the adopted guideline values and direct toxicity ecological benchmarks do not include consideration of exposure through the food chain which may result in bioaccumulation, biomagnification or exposures via multiple pathways. Therefore, it is necessary to assess the risks to higher trophic level organisms which may consume organisms that are directly exposed to PFAS in the various environmental media and which may have bioaccumulated PFAS in their tissues. The uptake through diet (food chain), as well as the direct ingestion of PFAS in surface water (drinking water) and the incidental ingestion of soil/sediment, has been included as part of dietary exposures in selected ROC (Section 5.2).

Risks associated with multiple potentially complete exposure pathways for ROC (detailed in Section 4.7) have been assessed using food web modelling methods derived from the US EPA. The food web model calculations are presented in Appendix N.

To evaluate the potential for toxic effects associated with PFOS / PFOA intake into upper trophic level organisms we have compared that total dietary intakes concentrations (doses) to the adopted toxicity



84

values, as detailed in Section 5.3. The dietary intake doses have been compared to the following toxicity values to indicate the likelihood and consequence of potential adverse effects:

• No Observed Adverse Effect Level (NOAEL): Dose at which no observable toxic effects were reported in laboratory studies and therefore interpreted that adverse effects have a low likelihood of occurring in the environment.

• Lowest Observed Adverse Effect Level (LOAEL): Lowest dose at which minor sub-lethal observable toxic effects were reported in any specimen in laboratory studies and therefore interpreted that toxic effects may potentially occur in the environment. Within a toxicity study this is a conservative toxicity value and does not represent lethal endpoints or effects observed to a large percentage of the test specimen.

• Tolerable Daily Intake (TDI): Conservative screening level toxicity value used for the derivation of the Canadian Wildlife Diet guideline values. Represents a higher reliability dose below which no observable toxic effects are expected to occur in the environment.

Due to the relatively limited toxicity data available for PFOS and PFOA towards upper trophic level organisms, it is recognised that there is uncertainty regarding the dose at which unacceptable toxic effects may occur. However, for the purpose of this ERA and to inform the risk management actions for the Base and surrounding area, we have interpreted the toxicity values in the following manner:

• Where the estimated dose is below the TDI it is unlikely that adverse toxic effects will occur. (Negligible Risk);

• Where the estimated dose exceeds the TDI but does not exceed the NOAEL there is a very low likelihood that adverse toxic effects will occur. (Very Low Risk);

• Where the estimated dose exceeds the NOAEL but does not exceed the LOAEL, there is a low likelihood that adverse toxic effects will occur. (Low Risk); and

• Where the estimated dose exceeds the LOAEL there is potential for unacceptable sub-lethal toxic effects to occur to some individuals and further risk assessment and/or risk management is considered necessary. (Moderate Risk).

The food-web modelling results are primarily discussed in relation to PFOS as this is considered to be the primary risk driving PFAS compound. Food-web modelling was also conducted for PFOA. The total dietary intake (dose) of PFOA for all ROC were below the lowest adopted toxicity values, as summarised in Table 6-9, as such the results for PFOA are not presented further in this Section, and risk from PFOA specifically is negligible.

Table 6-9: Maximum estimated total dietary intake (dose) of PFOA for Terrestrial and Aquatic receptors of concern relative to adopted toxicity values

Maximum total dietary intake

(dose) of PFOA (mg/kg/bw-day)

Toxicity Value (mg/kg/bw-day)

TDI NOAEL LOAEL

Terrestrial birds 0.002 0.0077 0.077 0.77

Terrestrial mammals 0.002 0.0011 0.100 0.40

Terrestrial reptiles <0.001 0.0077 0.077 0.77

Aquatic birds 0.001 0.0077 0.077 0.77

Aquatic mammals 0.001 0.0011 0.100 0.40

Aquatic reptiles <0.001 0.0077 0.077 0.77



85

6.2.1. Estimated Risks to Terrestrial Receptors of Concern

The results of the food-web modelling for each receptor of concern are presented in Appendix N and summarised below for terrestrial birds, terrestrial mammals and terrestrial reptiles.

Terrestrial Birds The estimated total dietary intakes for PFOS for the different terrestrial bird ROC species are summarised in Figures 6-1 to 6-3 in relation to the adopted toxicity values.

As shown in Figures 6-1 to 6-3 none of the estimated total dietary intakes for terrestrial bird species exceeded the adopted LOAEL. However, the adopted NOAEL was exceeded for six of the terrestrial bird species in Zones A, B and/or D as summarised in Table 6-10.

For those scenarios where the total dietary intake of PFOS exceeded the NOAEL we conducted a further evaluation of the dietary items driving the potential risk, as summarised in Table 6-10 and Figures 6-4 to 6-7.

This identified that ingestion of PFOS via incidental soil ingestion and/or consumption of terrestrial invertebrates were the two primary risk driving pathways. In relation to carnivorous birds the ingestion of PFOS in birds, reptiles and/or amphibians were the primary risk driving pathways. It is acknowledged that there is uncertainty around the food web modelling for carnivorous birds due to an absence of measured data on concentrations in birds as a dietary item. As such the maximum PFOS concentration in reptiles was used as a surrogate for the bird concentration. It is uncertain if this has resulted in an under- or over-estimation of the risks to carnivorous birds. Furthermore, it has been conservatively assumed that the carnivorous birds consume all their dietary items within the defined ecological exposure areas. It is considered plausible that carnivorous birds could have a larger feeding range, hence, it is uncertain if the potential risk for carnivorous birds may be overestimated.

Figure 6-1: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Herbivorous terrestrial birds. (Note: A = Zone A, B = Zone B, C = Zone C, D = Zone D)



86

Figure 6-2: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Invertivorous / omnivorous terrestrial birds. (Note: A = Zone A, B = Zone B, C = Zone C, D = Zone D)

Figure 6-3: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Carnivorous terrestrial birds. (Note: A = Zone A, B = Zone B, C = Zone C, D = Zone D)



87

Table 6-10: Terrestrial bird receptors of concern identified to have an estimated total dietary intake (dose) exceeding the NOAEL Receptors of Concern Receptors of concern identified have total dietary intake of PFOS exceeding the NOAEL (as indicated by shading) /

Primary dietary item driving potential risk

Zone A Primary Dietary Item Driving

Potential Risk

Zone B Primary Dietary Item Driving

Potential Risk

Zone C Primary Dietary Item Driving

Potential Risk

Zone D Primary Dietary Item Driving

Potential Risk


Rose-Crowned Fruit Dove

N - VL - VL - VL -

Australasian Figbird VL - VL - VL - VL -

Magpie Goose N - VL - VL - VL -

Bar-Shouldered Dove N - L Soil VL - L Soil

Plume Whistling Duck N - VL - VL - VL -


Masked Lapwing L Terrestrial invertebrates, birds

L Soil VL - L Soil

Straw-necked Ibis VL - L Soil VL - VL -

Black-face Wood Swallow

L Terrestrial invertebrates

VL - VL - N -


Black-Shouldered Kite VL - VL - VL - VL -

Wedge –Tailed Eagle L Birds, reptiles & amphibians

N - N - VL -

Whistling Kite L Birds VL - VL - N -

Black Kite L Reptiles & amphibians

VL - VL - VL -

No shading – Negligible Risk Green – Very Low Risk Yellow – Low Risk Orange – Moderate Risk



88

Figure 6-4: Estimated dietary intake (dose) of PFOS by dietary item: Terrestrial birds – Zone A

Figure 6-5: Estimated dietary intake (dose) of PFOS by dietary item: Terrestrial birds – Zone B



89

Figure 6-6: Estimated dietary intake (dose) of PFOS by dietary item: Terrestrial birds – Zone C

Figure 6-7: Estimated dietary intake (dose) of PFOS by dietary item: Terrestrial birds – Zone D



90

Terrestrial Mammals

The estimated total dietary intakes for PFOS for the different terrestrial mammal ROC species are summarised in Figures 6-8 to 6-10 in relation to the adopted toxicity values. As shown in Figures 6-8 to 6-10 none of the estimated total dietary intakes for terrestrial mammal species exceeded the adopted LOAEL. However, the adopted NOAEL was exceeded for four of the terrestrial mammal species in Zones A, B and/or D as summarised in Table 6-11. For those scenarios where the total dietary intake of PFOS exceeded the NOAEL we conducted a further evaluation of the dietary items driving the potential risk, as summarised in Table 6-11 and Figures 6-11 to 6-14.

This identified that ingestion of PFOS via incidental soil ingestion and/or consumption of terrestrial invertebrates were the two primary risk driving pathways. In relation to carnivorous mammals the ingestion of PFOS in birds, reptiles and/or amphibians were the primary risk driving pathways. It is acknowledged that there is uncertainty around the food web modelling for carnivorous mammals due to an absence of measured data on concentrations in birds as a dietary item. As such the maximum PFOS concentration in reptiles was used as a surrogate for the bird concentration. It is uncertain if this has resulted in an under- or over-estimation of the risks to carnivorous mammals. Furthermore, it has been conservatively assumed that the carnivorous mammals consume all their dietary items within the defined ecological exposure areas. It is considered plausible that carnivorous birds could have a larger feeding range, hence, it is uncertain if the potential risk for carnivorous mammals may be overestimated.

Figure 6-8: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Herbivorous terrestrial mammals. (Note: A = Zone A, B = Zone B, C = Zone C, D = Zone D)



91

Figure 6-9: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Invertivorous / Omnivorous terrestrial mammals. (Note: A = Zone A, B = Zone B, C = Zone C, D = Zone D)

Figure 6-10: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Carnivorous terrestrial mammals. (Note: A = Zone A, B = Zone B, C = Zone C, D = Zone D)



92

Table 6-11: Terrestrial mammal receptors of concern identified to have an estimated total dietary intake (dose) exceeding the NOAEL

Receptors of Concern Receptors of concern identified have total dietary intake of PFOS exceeding the NOAEL (as indicated by shading) / Primary dietary item driving potential risk

Zone A Primary Dietary Item Driving Potential Risk

Zone B Primary Dietary Item Driving Potential Risk

Zone C Primary Dietary Item Driving Potential Risk

Zone D Primary Dietary Item Driving Potential Risk


Pale Field Rat VL - VL - VL - VL -

Agile Wallaby VL - VL - VL - VL -

Black Flying Fox VL - N - N - N -

Black-Footed Tree Rat VL - VL - N - VL -

Grassland Melomys VL - VL - VL - VL -


Short-Beaked Echidna VL - VL - VL - VL -

Common Rock Rat VL - L Soil VL - L Soil

Northern Brown Bandicoot

VL - VL - VL - VL -

Large Bent-Winged Bat L Terrestrial invertebrates VL - VL - VL -

Orange Leaf-Nosed Bat L Terrestrial invertebrates VL - VL - VL -


Northern Quoll L Birds, reptiles & amphibians

VL - VL - VL -

Dingo VL - VL - VL - VL - No shading – Negligible Risk Green – Very Low Risk Yellow – Low Risk Orange – Moderate Risk



93

Figure 6-11: Estimated dietary intake (dose) of PFOS by dietary item: Terrestrial mammals – Zone A

Figure 6-12: Estimated dietary intake (dose) of PFOS by dietary item: Terrestrial mammals – Zone B



94

Figure 6-13: Estimated dietary intake (dose) of PFOS by dietary item: Terrestrial mammals – Zone C

Figure 6-14: Estimated dietary intake (dose) of PFOS by dietary item: Terrestrial mammals – Zone D



95

Terrestrial Reptiles

The estimated total dietary intakes for PFOS for the different terrestrial reptile ROC species are summarised in Figures 6-15 to 6-16 in relation to the adopted toxicity values.

As shown in Figures 6-15 to 6-16 none of the estimated total dietary intakes for terrestrial reptile species exceeded the adopted LOAEL. However, the adopted NOAEL was exceeded for one of the terrestrial reptile species, being the brown tree snake (a carnivore), in Zone A. Consumption of PFOS in birds was the primary risk driving exposure pathway, as shown in Figure 6-17.

Figure 6-15: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Invertivorous / Omnivorous terrestrial reptiles. (Note: A = Zone A, B = Zone B, C = Zone C, D = Zone D)

Figure 6-16: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Carnivorous terrestrial reptiles. (Note: A = Zone A, B = Zone B, C = Zone C, D = Zone D)



96

Table 6-13: Terrestrial mammal receptors of concern identified to have an estimated total dietary intake (dose) exceeding the NOAEL

Receptors of Concern Receptors of concern identified have total dietary intake of PFOS exceeding the NOAEL (as indicated by shading) / Primary dietary item driving potential risk

Zone A Primary Dietary Item Driving Potential Risk

Zone B Primary Dietary Item Driving Potential Risk

Zone C Primary Dietary Item Driving Potential Risk

Zone D Primary Dietary Item Driving Potential Risk

Invertivorous / Omnivorous Terrestrial Reptiles

Bynoe’s gecko N - N - N - N -

Gilbert’s Dragon N - N - N - N -

Striped Rainbow Skink VL - N - N - N -


Floodplain Monitor VL - N - N - N -

Olive Monitor VL - N - N - N

Brown Tree Snake L Birds N - N - N - No shading – Negligible Risk Green – Very Low Risk Yellow – Low Risk Orange – Moderate Risk



97

Figure 6-17: Estimated dietary intake (dose) of PFOS by dietary item: Terrestrial reptile – Zone A

6.2.2. Estimated Risks to Aquatic Receptors of Concern The results of the food-web modelling for each aquatic receptor of concern are presented in Appendix N and summarised below for aquatic birds, aquatic mammals and aquatic reptiles.

Aquatic Birds The estimated total dietary intakes for PFOS for the different aquatic bird ROC species are summarised in Figures 6-18 to 6-19 in relation to the adopted toxicity values.

As shown in Figures 6-18 to 6-19 none of the estimated total dietary intakes for aquatic bird species exceeded the adopted LOAEL. However, the adopted NOAEL was exceeded for five of the aquatic bird species in the Freshwater portion of Rapid Creek.

For those scenarios where the total dietary intake of PFOS exceeded the NOAEL we conducted a further evaluation of the dietary items driving the potential risk, as summarised in Figure 6-20. This identified that consumption of PFOS in fish as the primary risk driving exposure pathway. Consumption of PFOS in birds, reptiles & amphibians was also identified as a risk driving pathway for some bird species including the Black necked stork, Eastern great egret and White-faced heron.



98

Figure 6-18: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Invertivorous / Omnivorous Aquatic Birds. (Note: LC = Ludmilla Creek, RC-F = Rapid Creek Freshwater, RC-E = Rapid Creek Estuary, R&S = Reichardt & Sadgroves creeks)

Figure 6-19: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Invertivorous / Omnivorous Aquatic Birds. (Note: LC = Ludmilla Creek, RC-F = Rapid Creek Freshwater, RC-E = Rapid Creek Estuary, R&S = Reichardt & Sadgroves creeks)



99

Figure 6-20: Estimated dietary intake (dose) of PFOS by dietary item: Aquatic Birds – Rapid Creek (Freshwater)

Aquatic Mammals

The estimated total dietary intakes for PFOS for the aquatic mammal evaluated, the Large Footed Mytotis, is summarised in Figure 6-21 in relation to the adopted toxicity values.

As shown in Figures 6-21 the estimated total dietary intake for the Large Footed Mytotis did not exceed the adopted LOAEL. However, the adopted NOAEL was exceeded for the Freshwater portion of Rapid Creek. As shown in Figure 6-22, consumption of PFOS in fish was the primary risk driving exposure pathway.

Figure 6-21: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Piscivorous Aquatic Mammals. (Note: LC = Ludmilla Creek, RC-F = Rapid Creek Freshwater, RC-E = Rapid Creek Estuary, R&S = Reichardt & Sadgroves creeks)



100

Figure 6-22: Estimated dietary intake (dose) of PFOS by dietary item: Aquatic Mammals – Rapid Creek (Freshwater)

Aquatic Reptiles

The estimated total dietary intakes for PFOS for the different aquatic reptile ROC species are summarised in Figures 6-23 to 6-24 in relation to the adopted toxicity values.

As shown in Figures 6-23 to 6-24 none of the estimated total dietary intakes for aquatic bird species exceeded the adopted LOAEL and NOAEL values.

Figure 6-23: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Herbivorous, invertivorous & omnivorous aquatic reptiles. (Note: LC = Ludmilla Creek, RC-F = Rapid Creek Freshwater, RC-E = Rapid Creek Estuary, R&S = Reichardt & Sadgroves creeks)



101

Figure 6-24: Estimated total dietary intake (dose) of PFOS relative to the adopted toxicity values – Piscivorous aquatic reptiles. (Note: LC = Ludmilla Creek, RC-F = Rapid Creek Freshwater, RC-E = Rapid Creek Estuary, R&S = Reichardt & Sadgroves creeks)

Figure 6-25: Estimated dietary intake (dose) of PFOS by dietary item: Aquatic Reptiles – Rapid Creek (Freshwater)



102

6.2.3. Potential Contribution of Other PFAS Risks to Higher Trophic Level Organisms

The potential risks to higher trophic level organisms has been based on exposure to PFOS and PFOA. However, as discussed in Section 4.2, other PFAS compounds have been detected in abiotic and biotic media, which may contribute to toxic effects, therefore the potential risks could be underestimated by not considering other PFAS compounds. As such this section provides a qualitative evaluation of the potential for other PFAS compounds to contribute to toxicity and the overall estimates of risk for different scenarios.

Other PFAS compounds constitute less than 15% of the total PFAS concentrations for the majority of the terrestrial and aquatic biota sampled. Even if the other PFAS compounds are of equivalent toxicity to PFOS, the contribution of the other PFAS to toxic effect is not considered to be significant (i.e. no greater than 15% variance), particularly given the other inherent uncertainties and conservatism in the food web modelling.

Other PFAS compounds constitute a greater proportion of total PFAS in terrestrial plants, freshwater plants, freshwater invertebrates and estuarine invertebrates. However, incorporation of toxicity associated with other PFAS in these dietary items is considered unlikely to alter the risk outcomes as these dietary items were not the primary risk driving exposure pathways for any of the scenarios evaluated.

6.2.4. Risk Interpretation – Food Web Modelling Food web modelling was conducted for multiple ROCs to provide an estimate of potential risks to species in different trophic levels that have the potential to be exposed to PFAS compounds through ingestion of environmental media and dietary sources. Based on a range of exposure assumptions that have been adopted in the food web modelling, risks to individual (indicator) ROC can be extrapolated to gauge potential risks to other receptors within each trophic level.

Potential uncertainties have been identified which are important to understand and may affect the interpretation of risk noted within the current ERA. The assumptions and uncertainties are described in Section 1.6 and Section 7.

Based on the comparison of calculated intake doses to toxicity values, the likelihood and consequence of potential adverse effects were evaluated and risk categorised as:

• Negligible – where the estimated dose is below the TDI it is unlikely that adverse toxic effects will occur and the risks are considered to be negligible, further risk assessment and/or risk management actions is not considered necessary.

• Very Low – where the estimated dose exceeds the TDI but does not exceed the NOAEL there is no evidence that adverse toxic effects would be likely to occur, and as such the risk is considered to be very low. Further risk assessment and/or targeted risk management is unlikely to be required, unless conditions change.

• Low – where the estimated dose exceeds the NOAEL but does not exceed the LOAEL, there is some potential for adverse toxic effects to sensitive species. The overall risk is consider to be low as effects are likely to be sub-lethal and not affect a large portion of the population. Further risk assessment and/or precautionary risk management should be considered.

• Moderate – where the estimated dose exceeds the LOAEL for one or more species within the trophic level there is potential for unacceptable sub-lethal toxic effects to occur to sensitive species. The overall risk is considered to be moderate. Further risk assessment and/or risk management is considered to be warranted.



103

• High – where the estimated dose exceeds the LOAEL by an order of magnitude for more than one species within the trophic level there is likely to be unacceptable toxic effects to the general population of the trophic level, and the overall risk is considered to be high. Further risk assessment and/or risk management is considered to be warranted.

The risk interpretation from the food web modelling is presented and discussed below for each of the trophic levels in each of the EEAs.



104

Table 6-1: Risk interpretation summary Terrestrial Fauna

Group Trophic Level Zone A Zone B Zone C Zone D Dominant Pathway

Birds Herbivorous Very Low Low Very Low Low Incidental ingestion of soil

Invertivorous/ Omnivorous Terrestrial

Low Low Very Low Low Ingestion of terrestrial invertebrates

Carnivorous Low Very Low Very Low Very Low Ingestion of birds and reptiles

Mammals Herbivorous Very Low Very Low Very Low Very Low NA

Invertivorous/ Omnivorous

Low Low Very Low Low Ingestion of terrestrial invertebrates and incidental ingestion of soil.

Carnivorous Low Very Low Very Low Very Low Ingestion of birds

Reptiles Invertivorous/ Omnivorous

Very Low Negligible Negligible Negligible NA

Carnivorous Low Negligible Negligible Negligible Ingestion of birds

Aquatic Fauna

Rapid Creek Freshwater

Rapid Creek Estuary

Ludmilla Creek Darwin Harbour (Sadgroves &

Reichardt creeks)

Birds Invertivorous/ Omnivorous

Low Negligible Very Low Negligible Ingestion of reptiles and amphibians

Piscovorous Low Very Low Very Low Negligible Ingestion of fish, reptiles and amphibians

Mammals Piscovorous Low Very Low Very Low Very Low Ingestion of fish

Reptiles Piscovorous Very Low Negligible Negligible Negligible NA

Herbivorous, Invertivorous, Omnivorous

Negligible Negligible Negligible Negligible NA



105


The overall characterisation of risk to herbivorous terrestrial birds was “Very Low” to “Low”.

• Estimated intakes of PFOS by herbivorous terrestrial birds that inhabit the investigation area exceeded the NOAEL, but not the LOAEL, for at least one species in Zones B and D. This indicates there is some potential that small numbers of sensitive species may potentially be affected.

• The estimated intake in Zones A and C exceeded the TDI but not the NOAEL, indicating a very low risk of adverse effects to any species in the group.

• The LOAEL was not exceeded in any zone and therefore adverse effects to multiple species is unlikely.

• The dominant risk driving pathway was ingestion of soil derived from hotspots in Zones B and D.


The overall characterisation of risk to herbivorous terrestrial mammals was “Very Low”.

• The estimated intakes of PFOS did not exceed the NOAEL in any Zone, indicating a very low likelihood of adverse effects to any species.


The overall characterisation of risk for invertivorous or omnivorous birds was “Low” for Zones A, B, and D and “Very Low” for Zone C.

• Estimated intakes of PFOS by invertivorous and omnivorous terrestrial birds that inhabit the investigation area exceeded the NOAEL for at least one species in Zones A, B, and D, which indicates that some individuals of sensitive species may potentially be affected by reproductive or developmental effects. None exceeded the LOAEL, indicating a low likelihood of significant adverse effects to multiple species.

• The estimated intake in Zone C exceeded the TDI but not the NOAEL, indicating a very low risk of adverse effects to any species in the group.

• The dominant risk driving pathway was ingestion of terrestrial invertebrates and fish in Zone A, and was a combination of ingestion of invertebrates and incidental ingestion of soil in other zones.


The overall risk characterisation for invertivorous terrestrial mammals was “Low” for Zones A, B, and D and “Very Low” for Zone C.

• The estimated intake in Zone A, B and D exceeded the NOAEL for more than one indicator species but did not exceed the LOAEL for any indicator species. This indicates there is some potential that small numbers of sensitive species may potentially be affected.

• Estimated intakes of PFOS by invertivorous and omnivorous terrestrial mammals in Zone C exceeded the TDI but not the NOAEL, which indicates a very low risk of any adverse effects to any species in this area.

• The dominant risk driving pathway in Zone A was ingestion of terrestrial invertebrates and the dominant risk driving pathway in Zones B and D was incidental ingestion of soil.



106

Invertivorous/Omnivorous Terrestrial Reptiles

The overall risk characterisation for invertivorous or ominivorous reptiles was “Negligible” to “Very Low” across all zones.

• In Zones B, C and D the estimated intake of PFOS was below the TDI, which is the conservative screening level toxicity value and indicates that adverse effects are unlikely to occur.

• In Zone A the estimated intake of PFOS was above the TDI but below the NOAEL, which indicates a very low risk of adverse effects.


The risk characterisation for carnivorous birds was ”Low” in Zone A, and “Very Low” in Zones B, C and D.

• The estimated intake of PFOS in Zone A exceeded the NOAEL for most species but did not exceed the LOAEL for any species, indicating some potential that some individuals of sensitive species may potentially be affected by reproductive or developmental effects.

• The estimated intake in Zones B, C and D exceeded the TDI but not the NOAEL, indicating a very low risk of adverse effects to any species in the group.

• The dominant risk driving pathway in Zone A was ingestion of birds, amphibians and reptiles.


The overall risk characterisation for carnivorous mammals is “Low” in Zone A, and “Very Low” in Zones B, C and D.

• The estimated intake of PFOS in Zone A exceeded the NOAEL for one indicator species but did not exceed the LOAEL for any indicator species, indicating some potential that small numbers of sensitive species may potentially be affected.

• The estimated intake in Zones B, C and D exceeded the TDI but not the NOAEL for any indicator species.

• The dominant risk driving pathway in Zone A was ingestion of birds.


The overall risk characterisation for carnivorous reptiles was “Low” for Zone A, and “Negligible” for Zones B, C and D.

• The estimated PFOS intake in Zone A exceeded the NOAEL in one indicator species, but none exceeded the LOAEL, indicating some potential for adverse effects to sensitive species.

• The estimated intake in Zones B, C and D were below the TDI, which is the conservative screening level toxicity value. This provides a high level of certainty that adverse effects are unlikely to occur.

• The dominant risk driving pathway in Zone A was ingestion of birds.



107

Invertivorous/Omnivorous Aquatic Birds

The overall risk characterisation for invertivorous and omnivorous aquatic birds was “Low” for the freshwater reach of Rapid Creek, and “Very Low” to “Negligible” for estuarine Rapid Creek, Ludmilla Creek and Darwin Harbour (Sadgroves and Reichardt creeks).

• The estimated intake of PFOS in freshwater Rapid Creek exceeded the NOAEL for most species, but did not exceed the LOAEL for any indicator species, indicating some potential for adverse effects to some individuals of sensitive species.

• The estimated intake in Ludmilla Creek exceeded the TDI but not the NOAEL, indicating no evidence that adverse effects were likely to occur to any species in the group, and representing very low risk.

• The estimated intakes in estuarine Rapid Creek as well as Reichardt and Sadgroves creek were below the TDI, which is the conservative screening level toxicity value. This provides a high level of certainty that adverse effects are unlikely to occur.

• The dominant risk driving pathway in freshwater Rapid Creek was ingestion of reptiles and amphibians, although ingestion of aquatic plants and fish were also significant contributors.

Piscivorous Aquatic Birds

The overall risk characterisation for piscivorous birds was “Low” for freshwater Rapid Creek, and “Negligible” to “Very Low” for estuarine Rapid Creek, Ludmilla Creek, and Reichardt and Sadgroves creek.

• The estimated intake in freshwater Rapid Creek exceeded the NOAEL for all indicator species, but did not exceed the LOAEL, indicating some potential for adverse effects in some individuals in sensitive species.

• The estimated intake in Ludmilla Creek, estuarine Rapid Creek exceeded the TDI but not the NOAEL, indicating a very low risk of adverse effects to any species in the group.

• The estimated intake in Reichardt and Sadgroves creek was below the TDI, which is the conservative screening level toxicity value. This provides a high level of certainty that adverse effects are unlikely to occur.

• The dominant risk driving pathway in freshwater Rapid Creek was ingestion of fish, and ingestion of reptiles and amphibians.

Piscivorous Mammals

The overall risk characterisation for piscivorous mammals was “Low” for freshwater Rapid Creek, and “Very Low” for estuarine Rapid Creek, Ludmilla Creek, and Reichardt and Sadgroves.

• The estimated intake in freshwater Rapid creek exceeded the NOAEL but did not exceed the LOAEL, indicating some potential that sensitive species may potentially be affected.

• The estimated intakes for estuarine Rapid Creek, Ludmilla Creek, and Reichardt and Sadgroves exceeded the TDI but not the NOAEL, indicating a very low potential to adverse effects to any species.

• The dominant risk driving pathway for freshwater Rapid Creek was ingestion of fish.



108

Piscivorous Reptiles

The overall risk characterisation for piscivorous reptiles was “Very Low” for freshwater Rapid Creek and “Negligible” for Ludmilla Creek, estuarine Rapid Creek, and Reichardt and Sandgroves Creek.

• The estimated intake in freshwater Rapid Creek exceeded the TDI but not the NOAEL, indicating a very low risk of adverse effects to any species.

• The estimated intakes for estuarine Rapid Creek, Ludmilla Creek and Reichardt and Sandgroves Creek were below the TDI, which is the conservative screening level toxicity value. This provides a high level of certainty that adverse effects are unlikely to occur.

Herbivorous/Invertivorous/Omnivorous Aquatic Reptiles

The overall risk characterisation for herbivorous, invertivorous and omnivorous aquatic reptiles was “Negligible” in all waterways.

• The estimated intake in all waterways was less than the TDI, which is the conservative screening level toxicity value and indicates that adverse effects are unlikely to occur.



109

7. Uncertainty and Variability Analysis Ecological risk assessments require a number of assumptions regarding site conditions, presence of specific ecological receptors, exposure parameters and potential effects due to exposure to contaminants of potential concern, as such they have an element of uncertainty. Site-specific sampling and parameters were included where available to reduce the uncertainties in the assessment. However, limitations and uncertainties are expected with respect to the management of the data used and assumptions made. The assumptions considered were generally conservative in nature, based on currently available information, to account for uncertainty in the assessment.

7.1. Limitations and Uncertainty assessment The data presented in this ERA and used to characterise the risk from PFAS to ecological receptors at the Base and surrounds has associated limitations and uncertainties. Below is a list of the key potential limitations and uncertainties and how they may have affected the outcome of this ERA.

• Targeted sampling – The environmental sampling of abiotic (i.e., soil, surface water, and sediment) and biotic (i.e., tissue, serum) sample collected were targeted and likely included the ‘worst case’ concentrations in environmental media. Abiotic media sampling focused on source areas and areas where transport of PFAS across media may occur. This is likely to have overestimated the exposure point concentrations because sample collection was biased towards the impacted areas rather than gaining a representation of typical concentrations across each EEA. Had non-biased sampling across non-source areas or away from areas that may have been impacted been undertaken the overall calculated exposure to receptors would have been representative of the average ecological exposure for each EEA, but may have underestimated more highly exposed sub-populations. It is likely this would result in overall lower estimates of risk for the upper trophic level assessment (i.e. food web modelling).

• Limited guideline values, toxicity Benchmarks and study toxicity values for PFAS – At the time this ERA was prepared, Australia had not adopted nationally accepted direct toxicity benchmarks or toxicity reference values for use in assessing PFAS risk. Using the current published literature, the direct toxicity benchmarks and toxicity values used to assess PFAS risk in this ERA, beyond generic Tier 1 guideline values, were considered appropriate. Both the direct toxicity benchmarks and toxicity values were based on conservative sub-lethal chronic effects to the most sensitivity species tested, based on the best currently available information. Based on the global interest in the potential ecological risk of PFAS, new and continued research is being published at a rapid pace. It is noted that future research may result in direct toxicity benchmarks and toxicity values that are more- or less-conservative and it is acknowledged that additional information may warrant the amendment of the results of this ecological risk in the future.

There was insufficient toxicity data available to select toxicity values for different groups of avian and mammalian species. Therefore, the lowest toxicity values for avian and mammalian species were adopted to represent all species in these trophic groups. It is reasonable to assume that different species in these trophic levels will have differing sensitivities to PFAS, hence the potential toxicity to all avian and mammalian species cannot be accurately predicted.

Reliable toxicity values were not available for reptiles and amphibians, therefore the avian toxicity values used in this ERA were used to interpret the potential risk to reptiles. Therefore, the risk potential to reptiles has a high uncertainty.

It is acknowledged that there are differing opinions amongst the scientific community regarding the reliability of toxicity values to be adopted for ERAs and the application of uncertainty factors. As such for the purpose of this ERA we did not adopted a single toxicity reference value for calculation of risk. Rather we presented the total doses from the food web modelling in relation to a range of different toxicity values (i.e. lowest study NOAEL, lowest study LOAEL, and conservative guideline values, which incorporate uncertainty factors) to indicate the relative risks for different scenarios / receptors and inform prioritisation of further risk assessment and/or risk management actions.



110

• Abiotic Exposure Point Concentrations – For this ERA, only detected concentrations were used to determine potential exposure point concentrations for abiotic media and the maximum concentrations detected were used to evaluate potential direct toxicity risks. When the 95% UCL average concentration was selected to represent the exposure point concentration this was calculated using only detected concentrations (i.e. samples with non-detectable concentrations were excluded).

For the evaluation of direct toxicity to lower trophic level organisms the use of only the detected concentrations of PFOS/PFOA, and the maximum concentration, would result in the most conservative estimate of potential risk. This is likely to overestimate risk for lower trophic level organisms that are mobile as they may inhabit both impacted and unimpacted areas, however, it will highlight that some areas are driving risk, and should be targeted for management.

For the assessment of risk to upper trophic level organisms, which are thought to be more mobile than lower trophic level receptors, the risk would be overestimated by only basing risk on the detected concentrations. For example, in relation to soil across the entire investigation area, PFOS was detected in 65% of the samples, thus by limiting the evaluation to only detected concentration, the risk will be overestimated because ROCs may be exposed to soil that is non-detect for PFOS some of the time. Freshwater and estuarine surface water resulted in similar detection frequency: 92% in freshwater surface water (combined wet and dry season used for aquatic risk assessment), and 59% for estuarine surface water (combined wet and dry season used for aquatic risk assessment).

• Tissue Exposure Point Concentrations (Food Concentrations) – To evaluate the potential bioaccumulation of PFAS by upper trophic level ROC, multiple trophic level tissues were measured including aquatic and terrestrial invertebrates, fish, aquatic and terrestrial plants, small mammals, and reptiles & amphibians (bird tissues were not sampled this is discussed later in this section).

As with the abiotic media concentrations, on detected concentrations in dietary sources were used to determine the exposure point concentrations for food / dietary items used for food web modelling (i.e. non-detects excluded from 95% UCL calculations). This is likely to overestimate the potential risk.

Some fish and aquatic invertebrate tissue samples were collected to support the human health risk assessment (Coffey 2018b) and may not have represented the entire portion that an ecological receptor may consume. For fish tissue used in the assessment of piscivorous ROC, 89 of 147 samples or 60% were fillet or flesh only samples that may have under-estimated the fish tissue exposure. For aquatic invertebrates used in the assessment of invertivorous ROC, 68 of 146 samples or 46% were from only a portion of the aquatic invertebrate (i.e., flesh, organs, or tail) and may have under-estimated the aquatic invertebrate concentration compared to whole body results. For mammals used in the assessment of carnivorous ROC, 10 of 17 samples or 58% were modelled whole body concentrations based on measured blood serum concentrations (8 of 10), organs only (1 of 10), or flesh only (1 of 10). As different portions of mammals may bioaccumulate different concentrations of PFAS, the combined data set is likely to reflect the average tissue consumed, but some uncertainty is acknowledged.

• Lack of Bird Tissue Samples – Multiple ecological ROC that were evaluated as part of this ERA include avian tissue as part of their diet. Bird tissue data was not collected as part of the biotic sampling for this ERA. As such reptile tissue data was used as a surrogate. Based on US EPA guidance, when reptile tissue is not collected, bird bioaccumulation factors may be used to model reptile tissue data. In this instance, since avian tissue data was not collected, the reptile tissue data was used as a surrogate. Use of reptile tissue data as a surrogate for avian tissue is likely to have resulted in an overestimation of the potential risk to ROCs with avian tissue in their diet, as reptile tissue results were very high compared to other biota tested. The assessment could be refined through testing of bird tissue, but there is limited value in refining the uncertainty.



111

• Area Use Factor – For the purpose of the food web modelling undertaken for this ERA we assumed an area use factor of 1. This assumes that the receptors forage all their prey from a single exposure area (i.e. 100% of dietary items from one exposure area). This may overestimate the risk potential for some receptors if they gather food from a larger area. It was beyond the scope of this ERA to conduct a detailed review of the foraging range of different ROCs of concern. It was generally considered reasonable, given the extent of contamination in creeks, drains and riparian fringes that many animals would source all foods from impacted areas, however it is also reasonable to expect that larger birds, mammals and fish could have foraging ranges that extend beyond the defined boundaries of the specific exposure areas evaluated and therefore have lower average exposure than that estimated.

• Body Weight – For many of the receptors modelled the range of body weights present at the investigation area is unknown. Therefore, the expected body weights from published sources and field biologists in the Northern Territory was used for the food web modelling. Where a range of body weights has been reported we adopted the minimum body weight in both that allometric equations for ingestion rates. Within the likely weight range, food intake was relatively insensitive to the selected value.

• Diet Composition – The diet compositions adopted in the food web modelling for the ROCs have been estimated based on information from literature and professional judgement by field biologists in the Northern Territory. This may result in some inaccuracy where the actual diet composition varies from that adopted, however variability between species and between individuals is somewhat addressed by the application of multiple ROCs within feeding guilds.

• Food and Water Ingestion Rates – Australian species-specific information for the rates of food and water ingestion were applied where available in the published literature.Food and water ingestion rates were modelled using allometric equations, based on body weight, sourced from Nagy (2001). The use of allometric equations to estimate food and water ingestion rates may not accurately predict actual food ingestion rates, but the range of intake for different ROCs is likely to provide sufficient indication of intakes across ecological receptors.

• Limited Toxicological Information for Other PFAS – Currently, the global focus on PFAS toxicity assessment surrounds PFOA and PFOS and there are limited studies currently available that focus on the potential toxicities and risk of other PFAS. Therefore, the use of assessment of risk for PFOA and PFOS to represent the assessment of risk to all PFAS is likely to underestimate cumulative risk from all PFAS compounds. However, there are insufficient studies to define whether other PFAS are more or less toxic than PFOA or PFOS. It is noted that approximately 85% of the total PFAS present in the various abiotic and biotic media is present as PFOS and PFOA, hence the evaluation of risk based on PFOS and PFOA is considered to have evaluated the majority of the risk driving PFAS. As part of the ERA we conducted an evaluation of other PFAS in different abiotic and biotic media to appraise the potential for cumulative risk associated with other PFAS. This identified that PFAS other than PFOS and PFOA represented a significant portion of total PFAS in surface water, terrestrial plants and aquatic invertebrates hence potential for underestimation of risk to ROC where this media represent dietary items. However, it was noted for the ROC considered that consumption of surface water, terrestrial plants and aquatic invertebrates were not the primary risk driving exposure pathways, hence contribution from the other PFAS is unlikely to alter the overall risk conclusion for higher trophic level animal assessment. Direct risk from surface water may be underestimated by up to 50% (assuming PFOS and PFHxS have similar ecological toxicity, based on similar structure).

• Targeted Habitat and Ecological Surveys Conducted – As part of this ERA we conducted a desktop assessment and targeted field surveys to determine the habitats and ecological receptors most likely to be present at the investigation area. The brief surveys of habitat and ecological receptors were conducted in conjunction with the abiotic and biotic sample collection for this ERA and provided valuable information regarding potential receptors, foods and habitats available in each EEA. More detailed surveys of habitats and ecological receptor behaviour could be applied to determine if nominated ROCs at risk are actually present in the investigation area and that the assumed feeding behaviours are reasonable.

Notwithstanding the above, the data set used and conclusions drawn are valid.



112

7.2. Parameter Sensitivity A qualitative sensitivity analysis of the parameter used for the estimates of risks to lower trophic organisms (i.e. direct toxicity evaluation) and upper trophic organisms (i.e. food web modelling) is presented in Table 7-1. This provides a qualitative evaluation of the confidence in the input parameters used for risk calculations and how variance to these values may influence estimates of risk. The confidence in the value is defined as follows:

• Low: Insufficient data set and limited information available from literature sources or no site specific information available.

• Medium: Small data set, appropriate site/receptor information from literature, or reasonable alternative value applied.

• High: Reasonable data set or information available specific to the site / receptor.

• Conservative: Limited site specific information available, therefore a maximum or conservative value was selected. Therefore, although the specific value may not be accurate, the confidence in the scenario being protective is high.



113

Table 7-1: Parameter sensitive analysis

Parameter Confidence in the input value

Impact of increase in value on the risk estimate

Abiotic and biotic media concentrations

Soil concentrations High Increase

Surface water concentrations High Increase

Sediment concentrations Medium Increase

Terrestrial plant concentrations Medium Increase

Freshwater aquatic plant concentrations Medium Increase

Estuarine aquatic plant concentrations Medium Increase

Terrestrial invertebrate concentrations Medium Increase

Freshwater aquatic invertebrate concentrations Medium Increase

Estuarine aquatic invertebrate concentrations Medium Increase

Fish concentrations Medium Increase

Mammal concentrations Medium Increase

Reptile concentrations Medium Increase

Amphibian concentrations Medium Increase

Bird concentrations Low Increase

Food web modelling inputs

Abiotic dietary composition Medium Increase

Biotic dietary composition Medium Increase

Food ingestion rates Medium Increase

Water ingestion rates Medium Increase

Body weight Medium Decrease

Home range / Area Use Factor Conservative -

Direct toxicity benchmarks - lower trophic level organisms

Soil – Terrestrial plants Medium Decrease

Soil – Terrestrial invertebrates Medium Decrease

Surface water – Freshwater plants Medium Decrease

Surface water – Freshwater fish Medium Decrease

Marine/Estuarine waters - Invertebrates Medium Decrease



114

Parameter Confidence in the input value

Impact of increase in value on the risk estimate

Marine/Estuarine waters – Fish Low Decrease

Marine/Estuarine waters – Plants Low Decrease

Toxicity values – Upper trophic organisms

Avian Low Decrease

Mammalian Medium Decrease

Reptile Low Decrease

Taken as a whole, the assumptions used in the ERA are considered to be conservative, based on the best currently available information, and adopt the precautionary principle in estimating risk. The ERA approach presented does not consider a full probabilistic estimate of risk (i.e. evaluation of all the permutations of each input value), but presents conditional estimates based on a number of assumptions regarding the presence of ecological receptors, exposure parameters and toxicity. Thus, it is necessary to specify the assumptions and uncertainties inherent in the ERA to place the risk estimates into perspective. Risk assessment methodologies reflect an iterative process of development and as such should be recognised that this exposure assessment and risk assessment are based on existing methodologies and their limitations which may be subject to change.



115

8. Conclusions This ecological risk assessment has been conducted for PFAS contamination related to activities at RAAF Base Darwin (the Base). The primary objective of the ERA was to assess the potential for Base derived PFAS to cause adverse effects to ecological receptors that inhabit ecological habitats present on the Base and areas surrounding the Base. This was undertaken to guide risk management measures to address PFAS contamination from the Base.

Based on the conceptual site model of contamination migration from confirmed source areas and review of the likelihood of the presence of ecological receptors, potential risks to lower and upper trophic level organisms were assessed in four different terrestrial habitat zones and four difference aquatic habitat zones.

The terrestrial habitat zones evaluated were:

• Zone A - northeast quadrant of the investigation area encompassing the upper portion of Rapid Creek including the northern and southern headwater tributaries to Rapid Creek. .

• Zone B - northwest quadrant of the investigation area encompassing the middle and lower portions of Rapid Creek and eucalyptus woodland on Base.

• Zone C - the southwest quadrant of the investigation area, encompassing the main operating part of the Base and the area draining to Ludmilla Creek.

• Zone D - the southeast quadrant of the investigation area and encompassing facilities in the southern part of the base, and the main airstrip. .

The aquatic habitat zones evaluated were:

• Freshwater reach of Rapid Creek - including the Marrara Swamp and Rapid Creek through Pandanus swamps and Monsoon Rainforest.

• Estuarine portion of Rapid Creek - including the permanent estuarine reaches of Rapid Creek, nominally downstream of Trower Road.

• Ludmilla Creek – including the permanent estuarine waterway that receives fresh stormwater and groundwater discharge from the surrounding suburbs (including the Base).

• Reichardt and Sadgroves creeks - which are tidal arms of Darwin Harbour which also receive fresh water run-off and stormwater from Winnellie and the southern portion of the Base.

The potential risks to ecological receptors of concern (ROC) associated with exposure to Base derived PFAS were evaluated for each of the defined terrestrial and aquatic zones (or ‘ecological exposure areas’). This included:

• Direct toxicity evaluation - the potential risk of direct toxicity to lower trophic level receptors including invertebrates, fish, and plants were evaluated by comparing, measured concentrations in abiotic media including soil, sediment, and surface water to adopted direct toxicity benchmarks for these biota types. This included evaluation against general guideline values as well as direct toxicity benchmarks for specific receptors types.

• Food web modelling - the potential risk to upper trophic level receptors including birds, mammals, and reptiles were evaluated by modelling PFAS intake to ROC via dietary consumption of impacted abotic and biotic media. The PFAS intake does were then compared to dose based toxicity values to indicate the potential for adverse effects.



116

Based on the assessment conducted and described in Section 6.0, the following conclusions are provided.

Direct Toxicity

The direct toxicity evaluation included comparison of measured concentrations in abiotic media to general benchmarks to indicate the potential for direct toxic effects and then comparison to direct toxicity benchmarks for specific receptors types to indicate the nature of any potential direct toxic effects (i.e. types of organisms likely to be affected).

The findings from the direct toxicity evaluation are summarised in Tables 8-1 and 8-2 and outlined below. These conclusions are based on evaluation of the maximum measured concentrations in the abiotic media.

• Shallow Soil: Potential risks to terrestrial receptors were identified for exposure to shallow soils in isolated areas around source areas in Zones B, C and D as indicated by exceedance of the general guideline values. Further evaluation against direct toxicity benchmarks indicated a potential risk for both terrestrial plants and terrestrial invertebrates for Zones B and D. In relation to Zone C the direct toxicity evaluation indicated potential risks to terrestrial plants but not terrestrial invertebrates. It was considered unlikely that direct toxic effects would occur in Zone A as the maximum concentrations of PFOS and PFOA were below the general guideline value.

• Freshwater Surface Waters: The evaluation of direct toxicity to freshwater aquatic plants, aquatic water column invertebrates and fish identified some potential risk in Rapid Creek as indicated by PFOS concentrations exceeding the general guideline value. Further evaluation indicated that any adverse toxic effects are likely to be minimal for sensitive species as the maximum PFOS concentration was below the specific direct toxicity benchmark for aquatic plants and water column invertebrates. Some fish in Rapid Creek may potentially be affected by developmental effects based on the maximum PFOS concentration reported above the adopted fish toxicity value.

• Estuarine Surface Waters: The evaluation of direct toxicity indicated some potential for direct toxicity effects as the maximum PFOS concentrations exceeded the general guideline value in each of the ecological exposure areas. Average concentrations in Reichardt and Sadgroves creeks did not exceed the guideline value, indicating a very low risk in these waters. Further evaluation identified potential direct toxicity risks to estuarine aquatic water column invertebrates for estuarine waters in Rapid Creek, Ludmilla Creek, Sadgroves Creek and Reichardt Creek. It was considered unlikely that direct toxicity effects to aquatic plants would occur in Rapid Creek, Ludmilla Creek, Sadgroves Creek and Reichardt Creek as the maximum concentrations of PFOS and PFOA were below the specific direct toxicity benchmarks for this receptor. The influence of cumulative effects from other PFAS compounds potentially raised the risk for fish in the upper section of Ludmilla Creek, and based on comparison against freshwater fish toxicity benchmark, there is some risk of developmental effects to sensitive species in Ludmilla Creek and the estuarine section of Rapid Creek.

• Contribution of other PFAS Compounds: It is recognised that PFAS compounds other than PFOS and PFOA have been detected in abiotic media sampled from the investigation area. As such a qualitative evaluation of the potential contribution of these other PFAS compounds to direct toxic risk was undertaken. This evaluation identified that any contribution of other PFAS compounds to toxicity was unlikely to alter the risk outcome due to the relatively low contribution of other PFAS compounds to the total concentration and that the magnitude of the increase in PFOS concentrations that would by need to change the risk outcome (conservatively assuming the other PFAS compounds have the same toxicity as PFOS) for those scenarios were a potential risk is not indicated based only on PFOS and PFOA.



117

Risks to Upper Trophic Levels

Risks to upper trophic level organisms were evaluated based on the results from food web modelling conducted using a range of indicator species. The estimated intake does of PFOS and PFOA from dietary sources were compared to dose based toxicity values from the literature, including the TDI (conservative screening level toxicity value), lowest NOAEL and lowest LOAEL. Based on the comparison of intake does to toxicity values the likelihood and consequence of potential adverse effects were evaluated and risk categorised as negligible, very low, low, moderate or high.

The findings from the risk evaluation to upper trophic level organisms is summarised in Table 8-1, and outlined below. These conclusions relate to indicator species evaluated, however can notionally be applied to other organisms with similar diets and habitats within the broad trophic levels evaluated.

• Herbivorous Terrestrial Birds: Receptors in this feeding guild were assumed to consume a diet consisting of mostly of plants and to be exposed to PFAS in surface soils through incidental ingestion. The overall characterisation of risk to herbivorous birds was Very Low for Zone A and Zone C and Low of Zone B and Zone D. In Zones B and D the estimated intake dose of PFOS exceeded the NOAEL but not the LOAEL for at least one species, indicating some potential for sensitive species to be affected. The dominant risk driving pathway was incidental ingestion of soil.

• Herbivorous Terrestrial Mammals: Receptors in this feeding guild were assumed to consume a diet consisting of mostly plants and to be exposed to PFAS in surface soils through incidental ingestion. The overall characterisation of risk for herbivorous terrestrial mammals was Very Low.

• Invertivorous and Omnivorous Terrestrial Birds: Receptors in this feeding guild were assumed to consume a diet consisting of mostly of terrestrial invertebrates, with some also consuming fish and plants. Exposure to PFAS through the incidental ingestion of soils was also modelled as part of the potential uptake. The overall risk characterisation was Very Low for Zone C and Low for Zones A, B and D. In Zones A, B and D the PFOS intake dose exceeded the NOAEL but not the LOAEL for at least one species, indicating some potential for adverse effects to some individuals of sensitive species. The dominant risk driving pathway was ingestion of terrestrial invertebrates and fish in Zone A and a combination of invertebrates and incidental ingestion of soil in the other zones.

• Invertivorous and Omnivorous Terrestrial Mammals: Receptors in this feeding guild were assumed to consume a diet consisting mostly of terrestrial invertebrates, with some also consuming plants, reptiles and amphibians, as well as mammals. Exposure to PFAS through the incidental ingestion of soils was also modelled as part of the potential uptake. The overall risk characterisation was Very Low for Zone C and Low for Zones A, B and D. The PFOS dose in Zones A, B and D exceeded the NOAEL but not the LOAEL for more than one indicator species, indicating some potential for some individuals of sensitive species to be affected. The dominant risk driving pathway was ingestion of terrestrial invertebrates in Zone A.

• Invertivorous and Omnivorous Terrestrial Reptiles: Receptors in this feeding guild were assumed to consume a diet consisting of mostly of terrestrial invertebrates, with some also consuming other reptiles or amphibians. Exposure to PFAS through the incidental ingestion of soils was also modelled as part of the potential uptake. The overall risk characterisation was considered to be Negligible for Zones B, C and D and Very Low for Zone A. The estimated PFOS dose in Zone A exceeded the TDI but was below the NOAEL, which indicates a very low risk of adverse effects.

• Carnivorous Terrestrial Birds: Receptors in this feeding guild were assumed to consume a diet consisting mostly of mammals, with some also consuming invertebrates, reptiles and amphibians, as well as fish. The overall risk characterisation was considered to be Very Low in Zones B, C and D and Low in Zone A. The estimated intake dose of PFOS in in Zones B, C and D exceeded the TDI but not the NOAEL, indicating a very low risk of adverse effects to species in the trophic group. The dominant risk driving pathway in Zone A was ingestion of birds, amphibians and reptiles.



118

• Carnivorous Terrestrial Mammals: Receptors in this feeding guild were assumed to consume a diet consisting of mostly of mammals, with some also consuming birds, reptiles and amphibians. The overall risk characterisation was considered to be Very Low in Zones B, C and D and Low in Zone A. The estimated intake does of PFOS in Zone A exceeded the NOAEL but not that LOAEL, for one of the indicator species, indicating some potential that sensitive species may be affected. The dominant risk driving pathway in Zone A was ingestion of birds. Reptile tissue was used as a surrogate for bird tissue and may have over-estimated the risk potential.

• Carnivorous Terrestrial Reptiles: Receptors in this feeding guild were assumed to consume a diet consisting of invertebrates, reptiles and amphibians, mammals and birds. The overall risk characterisation was considered to be Negligible for Zones B, C and D and Low for Zone A. The estimated PFOS intake dose for Zone A exceeded the NOAEL but not the LOAEL in one indicator species, indicating some potential for adverse effects to sensitive species. The dominant risk driving pathway in Zone A was ingestion of birds. Reptile tissue was used as a surrogate for bird tissue and may have over-estimated the risk potential.

• Invertivorous and Omnivorous Aquatic Birds: Receptors in this feeding guild were assumed to consume a diet consisting of a mostly aquatic plants and aquatic invertebrates, with some also consuming reptiles and amphibians, fish, and mammals. Exposure to PFAS through the incidental ingestion of sediments was also modelled as part of the potential uptake. The overall risk characterisation was considered to be Very Low for Rapid Creek Estuary and Reichardt and Sadroves creeks, Very Low for Ludmilla Creek and Low for freshwater Rapid Creek. The estimated intake dose of PFOS in freshwater Rapid Creek exceeded the NOAEL for most indicator species but the LOAEL, indicating some potential for adverse effects. The dominant risk driving pathway in freshwater Rapid Creek was ingestion of reptiles and amphibians, aquatic plants and fish.

• Piscivorous Aquatic Birds: The diet of birds in this group was primarily fish with smaller percentages of invertebrates, reptiles and amphibians, and plants also consumed. For some of the receptors evaluated in this group the incidental ingestion of sediment was another potential exposure route that was modelled with respect to uptake. The overall risk characterisation was Very Low for estuarine Rapid Creek, Ludmilla Creek, and Reichardt and Sadgroves creeks. The estimated intake dose of PFAS in freshwater Rapid Creek exceeded the NOAEL for all indicator species but not the LOAEL, indicating some potential for adverse effects. The dominant risk driving pathway in freshwater Rapid Creek was ingestion of fish, reptiles and amphibians.

• Piscivorous Aquatic Mammals: The single piscivorous aquatic mammal evaluated had a diet that consisted of fish and mammals. The overall risk characterisation was considered to be Very Low for estuarine Rapid Creek, Ludmilla Creek and Reichardt and Sadgroves creeks, and Low for freshwater rapid creek. The estimated intake dose of PFOS in freshwater Rapid Creek exceeded the NOAEL but not the LOAEL, indicating some potential for adverse effects. The dominant risk driving pathway for freshwater Rapid Creek was ingestion of fish.

• Piscivorous Aquatic Reptiles: The diet of reptiles in this group was primarily fish or reptiles and amphibians, with smaller percentages of invertebrates, mammals, and plants also consumed. For some of the receptors evaluated in this group the incidental ingestion of sediment was another potential exposure route that was modelled with respect to uptake. The overall risk characterisation was considered to be Negligible for estuarine Rapid Creek, Ludmilla Creek and Reichardt & Sadgroves creek, and Very Low for freshwater Rapid Creek. The estimated intake dose of PFOS in freshwater Rapid Creek exceeded the TDI but not the NOEL for more than one indicator species indicating a very low risk of adverse effects.

• Herbivorous, Invertivorous and Omnivorous Aquatic Reptiles: The diet of reptiles in this group was primarily plants and invertebrates, with a small percentage of mammals and fish also consumed. For receptors evaluated in this group the incidental ingestion of sediment was another potential exposure route that was modelled with respect to uptake. The overall risk characterisation was considered to be Negligible for all waterways, with the estimate PFOS intake doses being less the than TDI.



119

• Contribution of other PFAS Compounds: The potential risks to upper trophic level organisms has been based on exposure to PFOS and PFOA. In is recognised that other PFAS compounds have been detected in abiotic and biotic media, which may contribute to toxic effect associated with dietary exposures. As such a qualitative evaluation of the potential for other PFAS compounds to contributed to toxicity and alter the overall estimates of risk was undertaken. This evaluation identified that any contribution of other PFAS compounds to toxicity was unlikely to alter the risk outcome due to the relatively low contribution of other PFAS compounds to the total concentration or the dietary items comprising higher proportions of other PFAS were not considered to be related to primary risk driving exposure pathways.

Table 8-1: Summary of potential risks to upper trophic level organisms

Risk Level Receptor (Ecological Exposure Area)

Terrestrial Aquatic

Negligible Invertivorous / omnivorous reptiles (Zone B, C & D)

Carnivorous reptiles (Zones B, C & D)

Invertivorous / omnivorous birds (RC-E, R&S)

Piscovorous birds (R & S) Piscovorous reptiles (RC-FW, RC-E, LC, R & S) Herbivorous, invertivorous, omnivorous reptiles (RC-F, RC-E, LC, R & S)

Very Low Herbivorous birds (Zones A and C)

Invertivorous / omnivorous birds (Zone C)

Carnivorous birds (Zones B, C & D) Herbivorous mammals (Zones A, B, C & D)

Invertivorous / omnivorous mammals (Zone C)

Carnivorous mammals (Zones B, C & D)

Invertivorous / omnivorous reptiles

Invertivorous / omnivorous birds (LC)

Piscovorous birds (RC-E, LC)

Piscovorous mammals (RC-E, LC)

Low Herbivorous birds (Zones B & D) Invertivorous / omnivorous birds (Zones A, B & D)

Carnivorous birds (Zone A) Invertivorous / omnivorous mammals (Zones A, B & D)

Carnivorous mammals (Zone A)

Carnivorous reptiles Zone A)

Invertivorous / omnivorous birds (RC-FW)

Piscovorous birds (RC-FW)

Piscovorous mammals (RC-FW)

Moderate None None

High None None

Note: LC = Ludmilla Creek, RC-F = Rapid Creek Freshwater, RC-E = Rapid Creek Estuary, R&S = Reichardt & Sadgroves creeks)



120

Overall Risk

Based on the findings from both the direct toxicity evaluation and assessment of risks to upper trophic level organisms the following overall conclusions are provided regarding the risk potential in different parts of the investigation area.

Table 8-2: Summary of overall risks to ecology

Habitat zone Direct toxicity Exposure to PFAS from the foods they eat

Main pathway

From soil From water Birds Mammals Reptiles

Zone A (Marrara Swamp)

Negligible Low – Very Low

Low – Very Low

Low – Very Low

Ingestion of fish, insects and reptiles

Zone B (Base north)

Medium Low – Very Low

Low – Very Low

Negligible Ingestion of soil and insects

Zone C (Base southeast)

Low Very Low Very Low Negligible

Zone D (Base south)

Medium Low – Very Low

Low – Very Low

Negligible Ingestion of soil

Rapid Creek Freshwater

Medium Low Low Negligible – Very Low

Direct contact risk to fish Ingestion of fish

Rapid Creek Estuary Low Negligible – Very Low

Very Low Negligible Direct contact risk to crustaceans and molluscs

Ludmilla Creek Low Very Low Very Low Negligible Direct contact risk to fish, crustaceans and molluscs

Darwin Harbour

Negligible Negligible Very Low Negligible

Terrestrial and riparian habitat along freshwater Rapid Creek (Zone A and banks of Freshwater Rapid Creek)

Low risk of adverse effects to sensitive species of upper trophic level animals was identified in Zone A (invertivorous, ominivorous and carnivorous birds and mammals) and along the freshwater reach of Rapid Creek (piscivorous birds and mammals). Direct toxicity to plants and lower trophic level terrestrial animals was a negligible risk, indicating that bioaccumulation through the food web in this area is increasing exposure. The adverse effects most likely to occur, based on toxicological studies used to derive toxicity values, are reproductive effects including reduced hatching rates and reduced fertility. Source areas of PFAS contamination are not present within this area and contamination is entering the area through surface water run-off and groundwater seepage from source areas to the south on the Base and potentially DIA.

Negligible Risk

The estimated PFAS exposure is below the conservative screening value, and unlikely to have adverse environmental effects.

Very Low Risk

The concentration or estimated intake exceeds the conservative screening value but there are no observed adverse effects in laboratory studies at those concentrations. There is a very low likelihood of adverse environmental effects.

Low Risk The concentration or estimated intake exceeds the conservative screening value and the reported no-observed-adverse-effect-levels from laboratory studies, but does not exceed the lowest concentrations where adverse effects were reported in laboratory studies. There is a low likelihood of adverse environmental effects, however observable effect cannot be excluded to some individuals or sensitive species.

Medium The concentration or estimated intake exceeds the lowest reported level that resulted in an adverse effect in laboratory studies. There is potential for adverse effects to some species in the associated habitat. Effects are not likely to be lethal or affect a majority of the ecological community.



121

Terrestrial habitats on Base (Zones B, C and D)

Low risks of adverse effects to plants and invertebrates have been identified on Base from isolated soil in source areas in Zones B, C and D. Elevated source areas driving the risk from soil contact are: former Fire Training Ground 1; former Fuel Farms 4 and 6; Hangar 31 and the current Fire Training Ground. Potential risk to sensitive species of upper trophic level animals (herbivorous and invertivorous birds and mammals) was also identified in these areas due to incidental ingestion of soil or invertebrates. Risks across these zones would be substantially reduced by limiting access to contaminated soils in source areas (current Fire Training Area, former Fire Training Area 2, Hangar 31, former Fuel Farms 4&6 and former Fire Training Area 1).

Aquatic habitat of freshwater Rapid Creek

Some potential risk via direct toxicity was identified to freshwater fish in Rapid Creek, potentially causing developmental effects based on the maximum PFOS concentrations.

Aquatic habitat of estuarine Rapid Creek

Risks from direct toxicity or bioaccumulation of PFAS in the estuarine section of Rapid Creek were limited to direct toxicity to aquatic invertebrates and potentially fish from contact with contaminated water. The likelihood of sustained adverse effects is considered low as the concentrations in this area were typically low, with only occasional elevated concentrations, likely during first flush or low tide events.

Aquatic habitat of Ludmilla Creek

Risks from direct toxicity or bioaccumulation of PFAS in Ludmilla Creek were limited to direct toxicity to aquatic invertebrates and potentially fish, based on maximum and average PFOS water concentrations. Risks are predominantly related to the section of Ludmilla Creek east of Dick Ward Drive, where elevated concentrations of PFOS in water were consistently reported as a result of stormwater and groundwater seepage into the upper reaches.

Aquatic habitat of Darwin Harbour (including Sadgroves Creek and Reichardt Creek)

Risks from direct toxicity or bioaccumulation of PFAS in the Darwin Harbour, as measured in Sadgroves Creek and Reichardt Creek, were low. Isolated zones of direct toxicity to sensitive aquatic invertebrates may occur in upper reaches where stormwater discharges occasionally introduce PFAS contamination.

These conclusions must be read in conjunction with the uncertainties and limitations of the assessment described in Section 7 and the attached Important Information about Your Coffey Report information (Appendix O).



122

9. References ATSDR. 2015. Toxicological Profile for Perfluoroalkyls. Agency for Toxic Substance and Disease Registry, U.S Department of Health and Human Services.

AECOM. 2011. Environment, Heritage and Risk Branch Initial Environmental Review - Final. September 19, 2011.

AECOM. 2016. Preliminary Ecological Risk Assessment RAAF Base Williamtown Stage 2B Environmental Investigation. Prepared for Department of Defence. December 9, 2016.

AECOM. 2017. Stage 2C Environmental Investigation Human Health Risk Assessment Army Aviation Centre Oakey, Environmental Investigation. Prepared for Department of Defence, 1 December 2017. Batley, G.E, van Dam, R.A., Warne, M. St. J, Chapman, J.C, Fox, D.R, Hickley, C.W and Stauber, J.L. (2014) Technical Rationale for Changes to the Method for Deriving Australian and New Zealand Water Quality Guideline Values for Toxicants. Prepared for the Council of Australian Government’s Standing Council on Environment and Water (SCEW).

Beyer, W. N., M.C. Perry, and P.C. Osenton. 2008. Sediment ingestion rates in waterfowl (Anatidae) and their use in environmental risk assessment. Integrated Environmental Assessment and Management 4(2):246-251.

Coffey (2018a) Detailed Site Investigation Report – Per- and Poly-fluoroalkyl Substances (PFAS), RAAF Base Darwin, Dated 5 February 2018

Coffey (2018b) Human Health Risk Assessment for PFAS, RAAF Base Darwin, Dated 19 June 2018

Department of the Environment and Energy (DoEE). 2016. Draft Commonwealth Environmental Management Guidance on Perfluorooctane Sulfonic Acid (PFOS) and Perfluorooctanoic acid (PFOA), October 2016.

Dietz, R., Gustavson, K., Sonne, C., Desforges, J-P., Riget, F.F., Pavlova, V., McKinney, M.A., Letcher, R.K. 2015. Physiologically-based pharmacokinetic modelling of immune, reproductive and carcinogenic effects from contaminant ecposure in polar bears (Ursus maritimus) across the Arctic. Environmental Research. 140 : 45 – 55.

EcoLogical Australia (ELA). 2014. Ecological Assessment, Proposed Flying Operations of the F-35 Aircraft. Prepared for Coffey. March 2014.

Environmental Agency (2004) Environmental Risk Evaluation Report: Perfluorooctane sulphonates (PFOS).

Environment Canada. 2006. Ecological Screening Assessment Report on Perfluorooctane Sulfonate, Its Salts and Its Precusors that Contan the C8F17S02 or S8F17S03, or C8F17SO2N Moiety.

Environment Canada. 2013. Perfluoroctane sulfonate in the Canadian environment. Environmental monitoring and surveillance in support of the chemicals management plan.

Environment and Climate Change Canada (ECCC). 2017. Canadian Environmental Protection Act, 1999, Federal Environmental Quality Guidelines, Perfluoroctane Sulfonate (PFOS). February 2017.

Environment and Climate Change Canada (ECCC). 2018. Canadian Environmental Protection Act, 1999, Federal Environmental Quality Guidelines, Perfluoroctane Sulfonate (PFOS). May 2018.

Hancock, P.J and Boulton, A.J. 2008. Stygofauna biodiversity and endemism in four alluvial aquifers in eastern Australia. Invertebrate Systematics. 22 : 117 – 126.

HEPA (2018) PFAS National Environmental Management Plan, January 2018. Heads of EPAs Australia and New Zealand (HEPA)



123

Hobbs, D.A, Warne, M St. J and Markich, S.J. (2005) Evaluation of Criteria Used to Assess the Quality of Aquatic Toxicity Data. Int. Env. Ass and Man. Vol(13) : 174 – 180.

Humphreys, W.F. 2000. The hypogean fauna of the Cape Range peninsula and Barrow Island northwestern Australia: In Wilkens, H, Culver, D.C, Humphreys, W.F. (Ed). Ecosystems of the World, Vol 30: Subterranean Ecosystems, Amsterdam, Elsevier.

Humphreys, W.F. 2008. Rising from Down Under developments in subterranean biodiversity in Australia from a groundwater fauna perspective. Invertebrate Systematics, 22 : 85 - 101

Kannan, K., Tao, L., Sinclair, E., Pastva, S.D., Jude, D.J., Giesy, J.P. 2005. Perfluorinated compounds in aquatic organisms at various trophic levels in a Great Lakes food chain. Arch. Environ. Contam. Toxicol. 48, 559 – 566.

Nagy, K.A. 2001. Food requirements of wild animals: Predictive equations for free-living mammals, reptiles, and birds. Nutrition Abstracts and Reviews, Series B 71:21R-32R.

National Environment Protection Council (NEPC). 2013. National Environment Protection (Assessment of Site Contamination) Measure 1999, as amended in 2013.

Newstead, J.L, Jones, P.D, Coady, K, Giesy, J.P. 2005. Avian Toxicity Reference Values for Perfluorooctane Sulfonate. Environ Sci Technol. 39 : 9357 – 9362.

Norwegian Pollution Control Agency (NPCA). 2008. Screening of Polyfluorinated Organic Compounds at Four Fire Training Facilities in Norway.

Sample, B.E., Opresko, D.M., Suter II, G.W. 1996. Toxicological Benchmarks for Wildlife: 1996 Revision

Sample, B.E., Aplin, M.S., Efroymoson, R.A., Suter II, G.W., Welsh, C.J.E. 1997. Methods and Tools for Estimation of the Exposure of Terrestrial Wildlife to Contaminants. Oak Ridge National Laboratory, October 1997.

Stahl, T., Matterm, D., Brunn, H. 2011. Toxicology of perfluorinated compounds. Environmental Sciences Europe. 23 : 28.

RIVM 2010. Environmental risk limits for PFOS. A proposal for water quality standards in accordance with the Water Framework Directive. Dutch National Institute for Public Health and the Environment. Report 601714013/2010.

U.S. Environmental Protection Agency (U.S. EPA). 1993. Wildlife Exposure Factors Handbook. Volume I of II. EPA/600/R-93/187. Office of Research and Development.

U.S. EPA (1994) AQUIRE (Aquatic toxicity information retrieval). United States Environment Protection Agency Office of Research and Development, Duluth, Minnesota, USA.

U.S Environmental Protection Agency (U.S EPA). 1997. Ecological Risk Assessment Guidance for Superfund: Process for Designing and Conducting Ecological Risk Assessments – Interim Final EPA 540-R-97-006. June 5, 1997.

U.S Environmental Protection Agency (U.S EPA). 1998. Guidelines for Ecological Risk Assessment EPA-630-R65-002 F, April 1998.

U.S EPA. 2004. Overview of the Ecological Risk Assessment Process in the Office of Pesticide Programs. Office of Prevention, Pesticides and Toxic Substances. Office of Pesticide Programs, Washington, D.C. January 23 2004.

U.S. EPA. 2016. ProUCL Version 5.1. Statistical Software for Environmental Application for Data Sets with and without Nondetect Observations. Updated 6/20/16.

WA EPA. 2013. Consideration of subterranean fauna in environmental impact assessment in WA, Environmental Assessment Guidance No. 12, June 2013, Environmental Protection Authority.



124

Weir, S.M, Yu, S, Know, A, Talent, L.G, Monks, J.M. and Salice, C.J. (2016) Acute toxicity and risks to lizards of rodenticide and herbicides commonly used in New Zealand, New Zealand Journal of Ecology. 40(3) : 342 – 350.

department of defence raaf base darwin · v2 revised draft 17/09/2018 aw sr sr v3 revised draft...

Documents