Perdaman Urea Project Cardno (WA) Pty Ltd

Air Quality Impact Assessment Final | Revision 7 16 March 2020 Air Quality Impact Assessment

Perdaman Urea Project

Project No: IW213400 Document Title: Air Quality Impact Assessment Document No.: Final Revision: Revision 7 Date: 16 March 2020 Client Name: Cardno (WA) Pty Ltd Project Manager: Lisa Boulden Author: Matthew Pickett, Maria Murphy & Andrew Boyd File Name: Perdaman-AQ-Assessment-Rev7_issued

Jacobs Group (Australia) Pty Limited ABN 37 001 024 095 Level 6, 30 Flinders Street Adelaide SA 5000 Australia T +61 8 8113 5400 F +61 8 8113 5440 www.jacobs.com

© Copyright 2020 Jacobs Group (Australia) Pty Limited. The concepts and information contained in this document are the property of Jacobs. Use or copying of this document in whole or in part without the written permission of Jacobs constitutes an infringement of copyright.

Limitation: This document has been prepared on behalf of, and for the exclusive use of Jacobs’ client, and is subject to, and issued in accordance with, the provisions of the contract between Jacobs and the client. Jacobs accepts no liability or responsibility whatsoever for, or in respect of, any use of, or reliance upon, this document by any third party.

Document history and status

Revision Date Description By Review Approved

A 12 Aug 2019 Preliminary draft M Pickett, M Murphy, A Boyd S Lakmaker, L Boulden L Boulden

B 6 Sep 2019 Draft report M Pickett, M Murphy, A Boyd S Lakmaker, L Boulden D Malins

0 26 Sep 2019 Draft report M Pickett, M Murphy, A Boyd L Boulden D Malins

1, 2 4 Oct 2019 Responses peer review M Pickett, M Murphy, A Boyd D Malins D Malins

3 24 Oct 2019 Changes after EPA review M Pickett, M Murphy, A Boyd L Boulden L Boulden

4 13 Nov 2019 Added CO re: EPA request M Pickett, M Murphy, A Boyd M Pickett L Boulden

5 6 Mar 2020 New modelling / report update M Pickett, M Murphy S Lakmaker D Malins

6 12 Mar 2020 Response to review comments M Murphy, M Pickett L Boulden L Boulden

7 16 Mar 2020 Updated following client review M Pickett L Boulden L Boulden

Final i Air Quality Impact Assessment

Contents Executive Summary ...... 1 1. Introduction ...... 5 1.1 Overview ...... 5 1.2 Scope ...... 5 1.3 Geographical Summary ...... 6 1.4 Sensitive Receptors ...... 7 2. Air Quality Standards ...... 9 2.1 Overview ...... 9 2.2 NEPM Standards for Criteria Pollutants ...... 9 2.3 Assessment Criteria for Ammonia, Formaldehyde and Methanol ...... 9 2.4 Air Quality Standards for Hydrocarbons ...... 10 2.5 Vegetation Protection Standards ...... 10 2.6 Land Surface Protection Standards ...... 11 3. Existing Air Quality ...... 12 3.1 Overview ...... 12

3.2 Airborne Particulate Matter: PM10 and PM2.5 ...... 12 3.3 Ammonia ...... 13 3.4 Nitrogen Dioxide and Ozone...... 14 3.5 Hydrocarbons...... 19 3.6 Sulfur Dioxide...... 19 3.7 Deposition Fluxes of Nitrogen and Sulfur ...... 19 4. Emissions Sources and Estimates ...... 22 4.1 Overview ...... 22 4.2 Risk Assessment...... 22 4.3 Model Scenarios ...... 25 4.4 Existing Emission Sources ...... 26 4.5 Future Emission Sources ...... 32 5. Modelling Methodology ...... 37 5.1 Overview ...... 37 5.2 Burrup Peninsula Modelling Applications ...... 37 5.3 TAPM Model configuration ...... 38 5.4 Modelled Variable Background Particulate Matter ...... 45 5.5 Deposition of Gaseous Pollutants ...... 46 5.6 Meteorological Modelling Performance ...... 47 6. Results ...... 48 6.1 Overview ...... 48

6.2 Particulate Matter as PM10 ...... 49

6.3 Particulate Matter as PM2.5 ...... 59

Final ii Air Quality Impact Assessment

6.4 Ammonia ...... 66 6.5 Nitrogen Dioxide ...... 69 6.6 Ozone ...... 76 6.7 Other Air Pollutants ...... 84 6.8 Potential Effects on Vegetation ...... 88 6.9 Deposition of Gaseous Pollutants ...... 93 6.10 Particulate Urea Deposition ...... 102 6.11 Summary of Results ...... 104 7. Comparisons with Deposition Monitoring ...... 114 8. Comparisons with Air Pollutant Monitoring...... 117 9. Conclusion ...... 120

Appendix A. Location Maps of Monitoring Stations...... 125 Appendix B. Local Meteorology ...... 127 Appendix C. Results – Meteorological Modelling ...... 142 Appendix D. Meteorological Modelling Performance ...... 144 Appendix E. Discrete Receptor Results – Air Quality ...... 145 Appendix F. Discrete Receptor Results – Vegetation ...... 154

List of Tables Table 1-1: Discrete Receptors for Air Quality Impact Assessment...... 7

Table 1-2: Discrete Receptors for Analysis of Model Results for Deposition ...... 8

Table 2-1: NEPM (Ambient Air) Standards relevant to the PU Project1...... 9

Table 2-2: NSW EPA Impact Assessment Criteria for Ammonia, Methanol and Formaldehyde ...... 10

Table 2-3: BTX Standards: NEPM (Air Toxics) MILs and NSW EPA Impact Assessment Criteria ...... 10

Table 2-4: EU (2008) Air Quality Standards for the Protection of Vegetation ...... 11

Table 3-1: Karratha Air Quality Monitoring – Data Capture NO2 and O3 ...... 16

Table 3-2: Dampier Air Quality Monitoring Results – Data Capture NO2 and O3 ...... 16

Table 3-3: Burrup Road Air Quality Monitoring Results – Data Capture NO2 and O3 ...... 16

Table 4-1: Summary of Current Air Emissions Sources Considered in the Modelling Assessment ...... 23

Table 4-2: Air Emissions Scenarios for Assessment ...... 25

Table 4-3: NWS Karratha Gas Plant Air Emissions Parameters ...... 27

Table 4-4: Pluto Onshore LNG Plant Air Emissions Parameters ...... 29

Table 4-5: Yara Pilbara Fertiliser and Yara Pilbara Nitrates TAN Air Emissions Parameters ...... 30

Final iii Air Quality Impact Assessment

Table 4-6: Yurralyi Maya Power Station Emissions Data ...... 30

Table 4-7: Devil Creek Gas Plant Air Emissions Parameters ...... 30

Table 4-8: West Kimberley Power Project Emissions Data ...... 31

Table 4-9: Karratha Power Station Emissions Data ...... 31

Table 4-10: Air Emissions Data for Shipping...... 31

Table 4-11: PU Project Air Emissions Sources and Parameters – Common Data ...... 32

Table 4-12: PU Project Air Emissions Sources and Air Emissions Estimates – Normal Operations ...... 33

Table 4-13: PU Project Air Emissions Sources and Air Emissions Estimates – Upset Conditions ...... 33

Table 4-14: Pluto LNG Development – Train 2 Air Emissions Parameters ...... 34

Table 4-15: Karratha Gas Plant – NWS Project Extension Proposal ...... 34

Table 4-16: Air Emissions Data for Methanol Proposal ...... 36

Table 5-1: Model Configuration ...... 39

Table 5-2: TAPM Vegetation Characteristics ...... 40

Table 5-3: Gaseous Dry Deposition Data Summary...... 47

Table 6-1: TAPM-GRS Results Presented as Contour Plots ...... 48

Table 6-2: Maximum 1-hour Average SO2 GLC (ppb)- Sensitive Receptors and Grid Maxima ...... 84

Table 6-3: Maximum 24-hour Average SO2 GLC (ppb)- Sensitive Receptors and Grid Maxima ...... 84

Table 6-4: Annual Average SO2 GLC (ppb)- Sensitive Receptors and Grid Maxima ...... 84

Table 6-5: Maximum Results for Annual NOx and SO2 GLC (ppb) ...... 88

Table 6-6: Summary of Results: Grid Receptor Maxima and Standards ...... 104

Table 6-7: Summary of Results: Grid Receptor Maxima and EU 2008 Standards ...... 104

Table 6-8: Summary of Results PNO: Discrete Receptor Maxima and Standards ...... 105

Table 6-9: Summary of Results PUC: Discrete Receptor Maxima and Standards ...... 106

Table 6-10: Summary of Results Baseline: Discrete Receptor Maxima and Standards ...... 107

Table 6-11: Summary of Results BPUC: Discrete Receptor Maxima and Standards ...... 107

Table 6-12: Summary of Results BPNO: Discrete Receptor Maxima and Standards ...... 108

Table 6-13: Summary of Results FPNO: Discrete Receptor Maxima and Standards ...... 109

Table 6-14: Summary of Results PNO: Discrete Receptor Maxima and EU 2008 Standards ...... 109

Table 6-15: Summary of Results Baseline: Discrete Receptor Maxima and EU 2008 Standards ...... 110

Final iv Air Quality Impact Assessment

Table 6-16: Summary of Results BPNO: Discrete Receptor Maxima and EU 2008 Standards ...... 110

Table 6-17: Summary of Results FPNO: Discrete Receptor Maxima and EU 2008 Standards ...... 110

Table 6-18: Summary of Results for Deposition: Grid Receptor Maxima and Medians ...... 110

2 Table 6-19: Comparisons of Model Results: NH3 Deposition (meq/m /year) ...... 111

2 Table 6-20: Comparisons of Model Results: NO2 Deposition (meq/m /year) ...... 112

2 Table 6-21: Comparisons of Model Results: SO2 Deposition (meq/m /year) ...... 113

Table 8-1: Comparisons of TAPM Results with 2014 Monitoring Results for Hourly Average NO2 ...... 118

Table 8-2: Comparisons of TAPM Results with 2014 Monitoring Results for Hourly Average O3 ...... 119

List of Figures Figure 1-1: PU Project Location in Industrial Area and Dampier and Karratha Urban Areas ...... 6

Figure 3-1: 24-hour average PM10 – Dampier Primary School 2001-2006 (Environmental Alliances 2007) ...... 13

Figure 3-2: Burrup Peninsula NH3 Monitoring Results 2016-2018 ...... 14

Figure 3-3: Locations of Meteorological and Air Quality Monitoring Stations ...... 15

Figure 3-4: Woodside Air Quality Monitoring Results 2009-2015: Karratha NO2 ...... 17

Figure 3-5: Woodside Air Quality Monitoring Results 2009-2015: Dampier NO2 ...... 17

Figure 3-6: Woodside Air Quality Monitoring Results 2009-2015: Burrup NO2 ...... 18

Figure 3-7: Woodside Air Quality Monitoring Results 2009-2015: Karratha O3 ...... 18

Figure 3-8: Woodside Air Quality Monitoring Results 2009-2015: Dampier O3 ...... 19

Figure 3-9: Total N & S Dry Deposition Flux 2004-2005 (Gillett, 2008) ...... 20

Figure 3-10: Total N & S Dry Deposition Flux 2007-2008 (Gillett, 2008) ...... 20

Figure 3-11: Summary of Total N & S Dry Deposition Flux 2003-2016 (YPN, 2017) ...... 21

Figure 4-1 Locations of Modelled Emissions Sources ...... 24

Figure 5-1: TAPM 3km and 1km Grids – Terrain, Vegetation and Land Use ...... 41

Figure 5-2: Deep Soil Moisture Content Settings ...... 42

Figure 5-3: CSIRO Biogenic NOx Area Emissions Database and Current Study Area (Inset) ...... 43

Figure 5-4: CSIRO Biogenic Area Rsmog Emissions Database and Current Study Area (Inset) ...... 44

Figure 5-5: Modelled Seasonally-Cyclic Background PM10 and Yara (2017) Measurements...... 46

3 Figure 6-1: PNO – Maximum 24h PM10 GLC (µg/m ) ...... 49

Final v Air Quality Impact Assessment

3 Figure 6-2: PUC – Maximum 24h PM10 GLC (µg/m ) ...... 50

3 Figure 6-3: Baseline – Maximum 24h PM10 GLC (µg/m ) ...... 51

3 Figure 6-4: BPNO – Maximum 24h PM10 GLC (µg/m )...... 52

3 Figure 6-5: BPUC – Maximum 24h PM10 GLC (µg/m ) ...... 53

3 Figure 6-6: FPNO – Maximum 24h PM10 GLC (µg/m ) ...... 54

3 Figure 6-7: PNO – Annual Average PM10 GLC (µg/m ) ...... 55

3 Figure 6-8: Baseline – Annual Average PM10 GLC (µg/m ) ...... 56

3 Figure 6-9: BPNO – Annual Average PM10 GLC (µg/m ) ...... 57

3 Figure 6-10: FPNO – Annual Average PM10 GLC (µg/m ) ...... 58

3 Figure 6-11: Baseline – Maximum 24h PM2.5 GLC (µg/m ) ...... 59

3 Figure 6-12: BPNO – Maximum 24h PM2.5 GLC (µg/m ) ...... 60

3 Figure 6-13: BPUC – Maximum 24h PM2.5 GLC (µg/m ) ...... 61

3 Figure 6-14: FPNO – Maximum 24h PM2.5 GLC (µg/m ) ...... 62

3 Figure 6-15: Baseline – Annual Average PM2.5 GLC (µg/m ) ...... 63

3 Figure 6-16: BPNO – Annual Average PM2.5 GLC (µg/m ) ...... 64

3 Figure 6-17: FPNO – Annual Average PM2.5 GLC (µg/m ) ...... 65

3 Figure 6-18: PNO – Maximum 1h NH3 GLC (µg/m ) ...... 66

3 Figure 6-19: PUC – Maximum 1h NH3 GLC µg/m ...... 67

3 Figure 6-20: Total Ammonia – Maximum 1h NH3 GLC µg/m ...... 68

Figure 6-21: Baseline – Maximum 1h NO2 GLC (ppb)...... 69

Figure 6-22: BPNO – Maximum 1h NO2 GLC (ppb) ...... 70

Figure 6-23: BPUC – Maximum 1h NO2 GLC (ppb) ...... 71

Figure 6-24: FPNO – Maximum 1h NO2 GLC (ppb)...... 72

Figure 6-25: Baseline – Annual Average NO2 GLC (ppb) ...... 73

Figure 6-26: BPNO – Annual Average NO2 GLC (ppb) ...... 74

Figure 6-27: FPNO – Annual Average NO2 GLC (ppb) ...... 75

Figure 6-28: Baseline – Maximum 1h O3 GLC (ppb) ...... 76

Figure 6-29: BPNO – Maximum 1h O3 GLC (ppb)...... 77

Figure 6-30: BPUC – Maximum 1h O3 GLC (ppb) ...... 78

Final vi Air Quality Impact Assessment

Figure 6-31: FPNO – Maximum 1h O3 GLC (ppb) ...... 79

Figure 6-32: Baseline – Maximum 4h O3 GLC (ppb) ...... 80

Figure 6-33: BPNO – Maximum 4h O3 GLC (ppb)...... 81

Figure 6-34: BPUC – Maximum 4h O3 GLC (ppb) ...... 82

Figure 6-35: FPNO – Maximum 4h O3 GLC (ppb) ...... 83

Figure 6-36: Baseline – Maximum 1h SO2 GLC (ppb) ...... 85

Figure 6-37: Baseline – Maximum 24h SO2 GLC (ppb) ...... 86

Figure 6-38: Baseline – Annual Average SO2 GLC (ppb) ...... 87

Figure 6-39: PNO – Annual Average NOX GLC (ppb) ...... 89

Figure 6-40: Baseline – Annual Average NOX GLC (ppb) ...... 90

Figure 6-41: BPNO – Annual Average NOX GLC (ppb) ...... 91

Figure 6-42: FPNO – Annual Average NOx GLC (ppb) ...... 92

2 Figure 6-43: PNO – Annual NH3 deposition (meq/m /year) ...... 94

2 Figure 6-44: Total Ammonia – Annual NH3 deposition (meq/m /year) ...... 95

2 Figure 6-45: Baseline – Annual NO2 deposition (meq/m /year) ...... 96

2 Figure 6-46: BPNO – Annual NO2 deposition (meq/m /year) ...... 97

2 Figure 6-47: FPNO – Annual NO2 deposition (meq/m /year) ...... 98

2 Figure 6-48: Baseline – Annual SO2 deposition (meq/m /year) ...... 99

2 Figure 6-49: BPNO – Annual SO2 deposition (meq/m /year) ...... 100

2 Figure 6-50: FPNO – Annual SO2 deposition (meq/m /year) ...... 101

Figure 6-51: PU Project Particulate Urea Deposition – Annual PM10 deposition (kg/ha/year) ...... 102

Figure 6-52: PU Project Particulate Urea Deposition – Annual PM2.5 deposition (kg/ha/year) ...... 103

2 Figure 7-1: NH3 Dry Deposition: Modelling and Monitoring (meq/m /year)...... 114

2 Figure 7-2: NO2 Dry Deposition: Modelling and Monitoring (meq/m /year) ...... 115

2 Figure 7-3: SO2 Dry Deposition: Modelling and Monitoring (meq/m /year)...... 116

Figure 7-4: Summations of N and S Dry Deposition: Modelling and Monitoring (meq/m2/year) ...... 116

Figure 9-1: Bureau of Meteorology and Air Quality Monitoring Stations ...... 125

Figure 9-2: Yara Pilbara Nitrates Air Quality Monitoring Stations (Yara, 2017) ...... 126

Final vii Air Quality Impact Assessment

Executive Summary

This report details the results of an air quality impact assessment to accompany environmental approvals processes required by the Western Australian Government for the Perdaman Urea Project (PU Project). The assessment included setting out detailed descriptions of the existing and potential future air pollutant emissions scenarios for the Burrup Peninsula. Air dispersion modelling was undertaken to determine how air pollutant emissions from all sources may impact on sensitive receptors on the Burrup Peninsula. The model predictions were assessed against air quality assessment standards, to gauge the current and potential future (cumulative) air quality impacts on human health and vegetation. Also, results are provided for the deposition of gaseous and particulate air pollutants, given these parameters may be of importance for the protection of Murujuga rock art on the Burrup Peninsula. Currently there are no standards for the protection of Murujuga rock art,

The CSIRO meteorological, air dispersion and photochemical model, ‘TAPM-GRS’, (The Air Pollution Model– Generic Reaction Set), was selected for modelling for reasons of reliability and efficiency. Air quality specialists of the Western Australian Government required that TAPM-GRS performance be verified as fit-for-purpose for the PU Project (and other similar Burrup Peninsula projects). In this undertaking the modelled results were compared with measurements from Woodside’s ambient air quality monitoring programs. Also, TAPM was used in the simpler ‘tracer’ or mass-dispersion mode for the assessment of some substances from the PU Project and other sources on the Burrup Peninsula.

The assessment included air dispersion modelling of particulate matter as PM10 and PM2.5, nitrogen dioxide (NO2), ozone (O3) and sulfur dioxide (SO2) for assessment against standards set out in the National Environmental Protection (Ambient Air Quality) Measure (‘NEPM’).

A supplementary assessment was undertaken for the deposition of urea, modelled as 100% of the PM10 emissions estimates for the PU Project.

Model results for ammonia (NH3), NO2 and SO2 deposition were provided specifically to support other assessments of the potential impacts to Murujuga rock art on the Burrup Peninsula. The assessment of impacts on rock art is outside the scope of this assessment, which would need to be completed by appropriately qualified and experienced rock art specialists.

Model results for airborne concentrations of oxides of nitrogen (NOx) and SO2 were obtained for comparison with European Union (2008) air quality standards for the protection of vegetation.

Key results for the PU Project’s air quality impact assessment were:

· TAPM-GRS results for airborne NO2 and O3 compared very well with Woodside’s measurements on the Burrup Peninsula indicating that the model was fit-for-purpose; i.e., hourly average statistics such as maximum, 99.9th percentile, 99th percentile, Robust Highest Concentration, typically lying within approximately 5 ppb of corresponding statistics for the measurements.

· There were no predicted exceedances of NEPM standards for NO2, O3, and SO2 concentrations for any of the emission scenarios that were investigated as part of this assessment. All results for these pollutants were well below relevant NEPM standards. · There were no predicted exceedances of ambient air quality standards (for the protection of human health) for NH3, formaldehyde and methanol; the latter two substances were eliminated from detailed assessment as low risk substances.

· There were no predicted exceedances of European Union (2008) air quality standards for NOx and SO2 for the protection of vegetation, for any of the air emissions scenarios.

· Estimates for dry deposition of gaseous NH3, NO2 and SO2 were determined from a combination of modelled results and calculations, which compared reasonably well with monitoring results for total dry deposition of nitrogen and sulfur-containing gases, adding further weight that the methods applied were fit-for-purpose.

Final 1 Air Quality Impact Assessment

· Examples of results for particulate urea deposition were: as PM10, 0.06 kg/ha/year; and the PM2.5 component, 0.002 kg/ha/year (90th percentiles of annual averages for 2601 grid receptor results).

In conclusion, based on modelling which showed compliance with relevant air quality criteria, there is a low risk of air quality impact on human health and vegetation from the PU Project. Model results for airborne concentrations of air pollutants compared well with monitoring. Model results for gaseous dry deposition compared well with monitoring, or typically within 5 meq/m2/year of monitoring results at eight sensitive receptor locations.

Final 2 Air Quality Impact Assessment

Important note about your report

The sole purpose of this report and the associated services performed by Jacobs is to provide air quality assessment services for the Perdaman Urea Project in accordance with the scope of services set out in the contract between Jacobs and the Client, Perdaman Chemicals & Fertilisers Pty Ltd.

In preparing this report, Jacobs has relied upon, and presumed accurate, any information (or confirmation of the absence thereof) provided by the Client and/or from other sources. Except as otherwise stated in the report, Jacobs has not attempted to verify the accuracy or completeness of any such information. If the information is subsequently determined to be false, inaccurate or incomplete then it is possible that our observations and conclusions as expressed in this report may change.

Jacobs derived the data in this report using various information sourced from Perdaman Chemicals & Fertilisers Pty Ltd, Woodside Energy Ltd and/or available in the public domain at the time or times outlined in this report. The passage of time, manifestation of latent conditions or impacts of future events may require further examination of the project, subsequent data analysis, and re-evaluation of the data, findings, observations and conclusions expressed in this report. Jacobs has prepared this report in accordance with the usual care and thoroughness of the consulting profession, for the sole purpose described above and by reference to applicable standards, guidelines, procedures and practices at the date of issue of this report. For the reasons outlined above, however, no other warranty or guarantee, whether expressed or implied, is made as to the data, observations and findings expressed in this report, to the extent permitted by law.

This report should be read in full and no excerpts are to be taken as representative of the findings. No responsibility is accepted by Jacobs for use of any part of this report in any other context.

This report has been prepared on behalf of, and for the exclusive use of Perdaman Chemicals & Fertilisers Pty Ltd and is subject to, and issued in accordance with, the provisions of the contract between Jacobs and Perdaman Chemicals & Fertilisers Pty Ltd. Jacobs accepts no liability or responsibility whatsoever for, or in respect of, any use of, or reliance upon, this report by any third party.

Final 3 Air Quality Impact Assessment

Abbreviations and Definitions

Abbreviation Expansion / Definition

ABS Australian Bureau of Statistics

BAAMP Burrup Ambient Air Monitoring Program

BoM Bureau of Meteorology

BSIA Burrup Strategic Industrial Area

CB Current Baseline Scenario

CSIRO Commonwealth Scientific and Industrial Research Organisation

EPA Environmental Protection Authority (Government of Western Australia)

FEED Front-End Engineering and Design

GLC Ground Level Concentration; an output from an air dispersion model commonly used for assessment

GRS Generic Reaction Set – a photochemical modelling scheme in-built to TAPM; e.g., see Hurley (2008a).

Jacobs Jacobs Group (Australia) Pty. Limited

KGP Karratha Gas Plant

LNG Liquefied Natural Gas

LPG Liquefied Petroleum Gas

MAC Murujuga Aboriginal Corporation

MLKC Murujuga Living Knowledge Centre

Mtpa Mega (million) tonne per annum

NEPM National Environment Protection Measure

NH3 Molecular formula for ammonia

NO Molecular formula for nitric oxide

NO2 Molecular formula for nitrogen dioxide

NOx Molecular formula for oxides of nitrogen, (NO and NO2)

NPI National Pollutant Inventory

O3 Molecular formula for ozone

PAQS Pilbara Air Quality Study

PCF Perdaman Chemicals & Fertilisers Pty Ltd

PLP Pluto on-shore LNG Plant

PM2.5 Particulate Matter 2.5 – mass concentration of particles with aerodynamic diameters less than 2.5 microns.

PM10 Particulate Matter 10 – mass concentration of particles with aerodynamic diameters less than 10 microns.

PU Project Perdaman Urea Plant

SIA (Burrup) Strategic Industrial Area

SKM Sinclair Knight Merz

SO2 Molecular formula for sulfur dioxide

TAN Technical Ammonium Nitrate (Yara Pilbara Nitrates)

TAPM The Air Pollution Model – a meteorological and air dispersion model developed by CSIRO (Hurley, 2008).

Tpd tonne per day

WEL Woodside Energy Limited

Final 4 Air Quality Impact Assessment

1. Introduction

1.1 Overview

Perdaman Chemicals and Fertilisers Pty Ltd (PCF) is proposing to establish a Urea Production Plant (PU Project) within the Burrup Strategic Industrial Area (BSIA), 7-8 km north-east (NE) from Dampier and 12-15 km north-west (NW) of the urban areas of Karratha on the north-west coastline of Western Australia (WA) (Figure 1-1). The PU Project intends to operate with a production capacity of approximately 2 million tonnes per annum (Mtpa) on Sites C and F within the BSIA. Natural gas for the PU Project will be sourced from a nearby domestic gas plant. The urea product will be transported via closed conveyor to the nearby Dampier Port for export via Panamax vessels (Cardno, 2019).

The PU Project consists of the following components: · Gas Supply Pipeline · Ammonia Plant · Urea Production Plant (Urea Plant) · Infrastructure and Logistics Building · Utility Block · Product Conveyor to Port · Product Storage Areas

The PU Project air emissions sources listed in Section 4.5.2 are the components of the PU Project relevant to the scope of this air quality impact assessment. Other components were excluded on the basis that they do not result in significant emissions to air. This air quality impact assessment, based on air pollutant dispersion modelling, was prepared to support applications for environmental approvals and to inform PCF of the potential impacts to air quality from the establishment of the PU Project.

In 2019, Woodside Energy Limited (WEL) engaged Jacobs to complete air quality assessments for several WEL projects in the Burrup Peninsula region along with development of an air pollutant dispersion model, which was an important foundation linking all these projects. A data sharing agreement was signed between PCF and WEL to allow for parts of the WEL air pollutant dispersion model to be used as a foundation for the PU Project.

1.2 Scope

This report provides an air quality impact assessment for the relevant components of the PU Project as defined in Section 1.1. The following items are within the scope of this report: · Modelling of the current air emissions scenario providing a ‘baseline’ to demonstrate current air quality on the Burrup Peninsula is good (consistent with monitoring data) and as determined by comparisons of the (expected) current air pollutant concentrations with air quality standards. · Modelling of air emissions associated with the proposed future operations of the PU Project in isolation. · Demonstration of cumulative air quality impacts associated with the best case, realistic worst case and most likely future emission scenarios for the Burrup Peninsula, in accordance with the (WA EPA 2016). This was done by assessment of the following air emissions scenarios: (1) PU Project–normal operations with Burrup Peninsula baseline (existing) emissions; (2) PU Project–upset conditions with baseline emissions; and (3) PU Project–normal operations with baseline and future proposed emissions.

Final 5 Air Quality Impact Assessment

1.3 Geographical Summary

The PU Project is proposed to be located on the Burrup Peninsula within site C and F of the BSIA on a lease area of 73.5 hectares. The proposed site is located 7-8 km NE of Dampier and 12-15 km NW of the urban parts of Karratha on the north-west coastline of Western Australia (WA) (Figure 1-1).

Figure 1-1: PU Project Location in Industrial Area and Dampier and Karratha Urban Areas

Final 6 Air Quality Impact Assessment

The Burrup Peninsula has significant cultural heritage value to Aboriginal people, particularly due to the large collection of rock art in the form of petroglyphs, standing stones, and other cultural sites such as foraging areas, ceremonial sites and hunting areas. The area is referred to by local indigenous groups as Murujuga and includes areas with protection as a National Heritage Place and National Park (DWER, 2019). The PU Project site is located adjacent to Murujuga National Park. Murujuga National Park is freehold land on the Burrup Peninsula, owned by the Murujuga Aboriginal Corporation (MAC).

The Australian Government transmitted a Tentative List Submission for the Murujuga Cultural Landscape to the United Nations Educational, Scientific and Cultural Organization (UNESCO) World Heritage Centre, with a view to adding area to Australia's World Heritage Tentative List later this year (WA Government, 2020).

1.4 Sensitive Receptors

The undertaking of air quality impact assessment using dispersion modelling normally includes a regular grid of receptors at ground-level, and the model results are assessed at those locations. The modelling grid used for this assessment is detailed in Section 5.3. However, the grid points may not be a good match for some sensitive receptor locations, so a separate set of so-called ‘discrete receptors’ that exactly match sensitive receptor locations may also be used. This approach was adopted for this assessment.

The discrete receptor representative locations used for the purpose of ambient air quality impact assessment, for the protection of human health, are listed in Table 1-1.

Table 1-1: Discrete Receptors for Air Quality Impact Assessment

Name MGA94 East (m) MGA94 North (m)

Woodside Air Quality Monitoring Station (AQMS), Burrup Road 476,665 7,721,038

Woodside AQMS Dampier 470,239 7,716,142

Woodside AQMS Karratha 484,892 7,707,575

Ngajarli (ex-‘Deep Gorge’) 477,964 7,718,020

Hearson Cove 478,928 7,718,358

Murujuga National Park - central northern extent 483,354 7,730,501

Murujuga National Park - central southern extent 476,195 7,714,869

Representative of MAC office, King Bay 475,574 7,719,459

Standing Stones 474,714 7,717,782

Murujuga Living Knowledge Centre (MLKC) 479,900 7,727,100

Discrete receptors were also used to extract model results for deposition at specific locations, for comparisons with measurements of various pollutant species, and to provide information about deposition that may be used by other assessment reports; these DRs are listed in Table 1-2. (Note rock art impact assessment is outside the scope of this assessment).

Specifically, the DRs listed from Dolphin Island to Mardie Station were used for comparisons with measurements from CSIRO monitoring at those locations; e.g., Gillett (2008).

The additional DRs listed from Hearson Cove to Murujuga Living Knowledge Centre (MLKC) were added by Cardno/Perdaman to widen the analysis of the model results for deposition.

Final 7 Air Quality Impact Assessment

Table 1-2: Discrete Receptors for Analysis of Model Results for Deposition

Name MGA94 East (m) MGA94 North (m)

Dolphin Island (CSIRO) 484,598 7,738,456

North Burrup (CSIRO) 482,347 7,730,288

Woodside East (CSIRO) 477,363 7,721,921

Burrup Road (CSIRO) 475,961 7,719,787

Water Tank (CSIRO) 477,616 7,720,114

Ngajarli (ex-‘Deep Gorge’) (CSIRO) 477,964 7,718,020

King Bay south (CSIRO) 474,026 7,717,213

Karratha (CSIRO) 482,990 7,707,089

Mardie Station (CSIRO) 408,643 7,659,017

Hearson Cove 478,928 7,718,358

Murujuga NP - central northern extent 483,354 7,730,501

Murujuga NP - central southern extent 476,195 7,714,869

Representative of MAC office, King Bay 475,574 7,719,459

Standing Stones 474,714 7,717,782

Murujuga Living Knowledge Centre (MLKC) 479,900 7,727,100

Final 8 Air Quality Impact Assessment

2. Air Quality Standards

2.1 Overview

This section sets out legislation, policy and guidelines applicable to air assessments in WA, relevant to the PU Project.

2.2 NEPM Standards for Criteria Pollutants

The WA Environmental Protection Authority (EPA) provides guidance for assessing the potential impacts of the PU Project on air quality in the Environmental Factor Guideline: Air Quality, published in 2016 (EPA, 2016); whilst this does not specify air quality standards for assessment it does provide the following considerations: · Whether numerical modelling and other analyses to predict potential impacts have been undertaken using recognised standards with accepted inputs and assumptions. · Whether existing background air quality, including natural variations, have been established through monitoring and accepted proxy data. · Whether analysis of potential health and amenity impacts have been undertaken using recognised criteria and standards, where relevant, informed by Australian and international standards.

In the absence of specific air quality standards from the EPA, it is common practice for the NEPM (Ambient Air Quality) to be adopted for air quality impact assessments in WA. To assess potential ground level concentrations (GLC) for the PU Project, modelled predictions were assessed against NEPM (Ambient Air Quality) standards shown in Table 2-1.

Table 2-1: NEPM (Ambient Air) Standards relevant to the PU Project1

Air pollutant Averaging period Maximum concentration Maximum allowable standard exceedances

Nitrogen dioxide (NO2) 1 hour 120 ppb 1 day a year 1 year 30 ppb None

Ozone (O3) 1 hour 100 ppb 1 day a year 4 hours 80 ppb 1 day a year

Sulfur dioxide (SO2) 1 hour 200 ppb 1 day a year 1 day 80 ppb 1 day a year 1 year 20 ppb None

3 Particulate Matter 10 (PM10) 1 day 50 µg/m None

1 year 25 µg/m3 None

3 Particulate Matter 2.5 (PM2.5) 1 day 25 µg/m None

1 year 8 µg/m3 None

2.3 Assessment Criteria for Ammonia, Formaldehyde and Methanol

The NSW EPA impact assessment criteria (NSW EPA, 2016), were used for the assessment of ammonia, methanol and formaldehyde emissions from the PU Project. The hourly average impact assessment criteria were the most practical standards to adopt for the modelled hourly average results for GLCs in the absence of standards specified by WA regulators (Table 2-2).

Final 9 Air Quality Impact Assessment

Table 2-2: NSW EPA Impact Assessment Criteria for Ammonia, Methanol and Formaldehyde

Pollutant 99.9th Percentile 1-hour average Maximum 1-hour average (ppb)1 (µg/m3)2

3 Ammonia (NH3) 460 ppb 330 µg/m

3 Formaldehyde (CH2O) 18 ppb 20 µg/m

Methanol 2400 ppb 3000 µg/m3

1. NSW EPA (2016) quotes the gas volumes (ppb) expressed at 25°C and at an absolute pressure of 1 atmosphere (101.325 kPa). 2. Level 2 impact assessment criteria are 99.9th percentile hourly averages. For this assessment the (more conservative) maximum hourly averages were compared with the criteria in each case. The maxima were still substantially less than the criteria.

2.4 Air Quality Standards for Hydrocarbons

Commonly the measurement suite benzene, toluene and xylenes (BTX) are used for air quality assessment of airborne hydrocarbons, or Volatile Organic Compounds (VOCs). The normal function of the National Environment Protection (Air Toxics) Measure 2011 is to assess VOCs monitoring data with matching (generally longer) averaging periods. Other guidance or policy documents that provide hourly average standards can be used for the assessment of model results for shorter term averages of individual VOCs such as benzene. The NSW Environment Protection Authority (EPA) Approved Methods for the Modelling and Assessment of Air Pollutants in NSW (NSW EPA, 2016), is an example.

The NEPM (Air Toxics) sets out Monitoring Investigation Levels (MILs) for the assessment of VOCs; MILs used for interpretation of BTX monitoring data for this assessment are listed in Table 2-3. The corresponding NSW EPA (2016) hourly average impact assessment criteria are also listed, for comparison.

Table 2-3: BTX Standards: NEPM (Air Toxics) MILs and NSW EPA Impact Assessment Criteria

NSW EPA (2016) impact assessment Pollutant NEPM MIL, averaging period criterion, averaging period (gas vol., 25oC)

Benzene 3 ppb, annual 9 ppb, 1 hour (max.)

1000 ppb, 24-hour Toluene 90 ppb, 1 hour (max.) 100 ppb, annual

250 ppb, 24-hour Xylenes 40 ppb, 1 hour (max.) 200 ppb, annual

Formaldehyde 40 ppb, 24-hour 18 ppb, 1 hour (max.)

2.5 Vegetation Protection Standards

Air quality standards for the protection of vegetation have been set out by the World Health Organization (WHO, 2000), and the European Union (EU, 2008). While these standards were developed for the protection of a variety of vegetation in the European region, such as conifer forests, they have had wider application and have been used for the assessment of similar projects in WA previously; e.g., SKM (2006) used WHO (2000) standards. This air quality impact assessment has adopted the EU (2008) standards for SO2 and oxides of nitrogen (NOx) given they are the most recent; the relevant standards are listed in Table 2-4.

Final 10 Air Quality Impact Assessment

Table 2-4: EU (2008) Air Quality Standards for the Protection of Vegetation

Equivalent Standard Adopted for Assessment; Annual Air Pollutant EU (2008) Air Quality Standard Average

3 o SO2 20 µg/m , annual 7.8 ppb at 30 C (typical Burrup daytime temp.)

3 o NOx 30 µg/m , annual 16.2 ppb at 30 C (typical Burrup daytime temp.)

Note: NOx means the sum of NO and NO2.

Air dispersion models calculate surface deposition for airborne substances using an airborne concentration near ground-level, a deposition velocity for the substance of interest, and other parameters; e.g., Seinfeld and Pandis (2016). The calculation of deposition leads to additional uncertainties, therefore the vegetation standards for deposition were not used for this assessment – the standards for airborne concentrations were used (Table 2-4).

2.6 Land Surface Protection Standards

There are no accepted or commonly applied standards for assessing deposition of air pollutants on land surfaces, such as Burrup Peninsula Aboriginal rock art. The Government of WA Murujuga Rock Art Strategy (2019) indicates further research is needed in this area. While this assessment report provides results for dry deposition of NH3, NO2 and SO2, no assessment nor commentary is provided about the potential impacts on rock art.

Final 11 Air Quality Impact Assessment

3. Existing Air Quality

3.1 Overview

The purpose of this section is to describe existing air quality in the Burrup Peninsula region, primarily by a review of Woodside ambient air quality monitoring data. To understand the existing meteorology relevant to the PU Project, a review of local meteorology on the Burrup Peninsula is provided in Appendix B.

Woodside established the Burrup Ambient Air Monitoring Program (BAAMP) in 2008, which continued to 2011. As part of the Pluto project, Woodside continued the monitoring program to the end of 2015. Prior to the more recent monitoring programs by Woodside, the Pilbara Air Quality Study (PAQS) was undertaken by the Government of Western Australia in the early 2000s including investigations of monitoring data. The PAQS established important baselines for future assessments (DEP, 2002; DoE, 2004).

In summary, the review of air quality monitoring data for the Burrup Peninsula study area showed that NO2, O3, PM10, and PM2.5 are higher risk air quality indicators. While NO2, O3 and SO2 concentrations have not exceeded NEPM (Ambient Air) standards, PM10 and PM2.5 concentrations have exceeded the NEPM (Ambient Air) standards on several occasions each year, most likely due to fires and dust storms.

3.2 Airborne Particulate Matter: PM10 and PM2.5

The particulate size classes of airborne Particulate Matter (PM), PM10 and PM2.5, are defined in the abbreviations section. The existing environment around the Burrup Peninsula is characterised by occasionally high levels of PM from events such as bushfires and storms; e.g., Air Assessments (2010b). Rio Tinto conducts PM monitoring at Dampier, Karratha, King Bay, Wickham, Point Samson and Roebourne (Rio Tinto, 2015), though monitoring reports were not available for review at the time of writing. Recent data are published online (Pilbara Iron, 2019) and show that, for example, on the 9 May 2019, very high PM10 concentrations were observed at Dampier, Karratha, Wickham, Point Samson, and Roebourne. The strong correlation between these measurements, taken by several monitors on this day, suggests a dust storm was the probable cause.

Environmental Alliances (2007) provided a useful time series plot of daily PM10 measured at Dampier by Hamersley Iron over 2001-2006 (Figure 3-1). Some broad conclusions about the variations in PM10 on the Burrup Peninsula can be drawn by inspection of this relatively long-term record. The record provides information about the clean-air background and air quality impacts, with the latter likely due to local particulate emissions from bushfires, dust storms, and some industry. The PM10 concentrations peaked during higher wind speeds in January, with typical daily concentrations ranging between 30-40 µg/m3. Exceedances of the NEPM (Ambient Air Quality) standard of 50 µg/m3 ranged from 5 to 10 exceedances per year. Mid-year, during the dry season 3 with lower wind speeds, typical daily concentrations varied between 10-20 µg/m . A review of 30 days of PM10 data for Karratha, (10 April to 10 May 2019), indicates the ‘clean air background’ PM10 levels are approximately 3 3 10 µg/m , with a median or average closer to approximately 20 µg/m . These values are typical of PM10 concentrations measured in other parts of Australia, such as Keywood et al. (2017).

The Pluto LNG Development Cumulative Air Quality Study (SKM, 2006) reviewed monitoring results for particulate matter as PM10. The study found that existing industrial activity in the Pilbara mainly contributed to emissions of PM2.5 and PM10, with exceedances of NEPM standards. SKM (2006) stated that higher PM10 concentrations were observed on days of high wind speeds. On these days the PM2.5/PM10 fraction was reduced from approximately 50% to approximately 20%, pointing to wind-blown dust as the cause of the higher PM10 concentrations rather than smoke emissions, which comprise more smaller particles.

The review by Air Assessments (2010a) indicated that measurements of PM10 at Dampier tend to be high, and “exceed the NEPM (Ambient Air Quality) standard”. Air Assessments (2010a) indicated the major sources of particulate matter in the Burrup region are: smoke from fires, dust from wind storms and off iron ore stockpiles, and ship-loading operations at the ports of Dampier and Cape Lambert.

Final 12 Air Quality Impact Assessment

Figure 3-1: 24-hour average PM10 – Dampier Primary School 2001-2006 (Environmental Alliances 2007)

3.3 Ammonia

Ammonia (NH3) is ranked third among the nitrogen-containing compounds in the atmosphere, after gaseous nitrogen and nitrous oxide. Over land, the main sources are animal waste, emissions from soils, and industrial emissions. Background NH3 concentrations in continental air (land sources) range from 0.1–10 ppb (Seinfeld and Pandis, 2016). Cattle feed lots are a significant source of higher NH3 concentrations within a radius of approximately 7 km from the feed lots; recent Australian examples are: Shen et al. (2016) and Hacker et al. (2016). Low-level airborne measurements of NH3 in Victoria by Hacker et al. (2016) showed background NH3 levels ranging from approximately 1 ppb near sunrise and sunset to approximately 2 ppb near midday.

3 o Gillett et al. (2012) determined a background NH3 level of 0.5 ppb (0.35 µg/m at 25 C) in their review of results from eight monitoring stations on the Burrup Peninsula obtained in 2004-2005. The general conclusion was concentrations of NH3 (and other pollutants) in the Burrup Peninsula region were similar to other remote terrestrial areas, and very low compared to polluted urban areas.

Strategen (2018) provided a summary of results of NH3 monitoring near the Yara Pilbara fertilisers and nitrate plants on the Burrup Peninsula over 2016-2018. A statistical summary of these results is shown plotted in Figure 3-2; the statistics are: maximum, average plus and minus one standard deviation; and average (the minima were all zero). The Gillett et al. (2012) estimate of 0.5 ppb (0.35 µg/m3 at 25oC) is also shown in the plot. The 2004-2005 results described by Gillett et al. (2012) provide a clearer estimate of the background NH3 levels for the Burrup region, unaffected by industrial and other sources.

Final 13 Air Quality Impact Assessment

Figure 3-2: Burrup Peninsula NH3 Monitoring Results 2016-2018 *The previous name for Ngajarli was Deep Gorge.

3.4 Nitrogen Dioxide and Ozone

NOx and O3 are key pollutants associated with the PU Project. Whilst NOx is a direct emission, the production of O3 is a more complex process; O3 is produced in the atmosphere from emissions of NOx and other pollutants such as VOCs and CO, in the presence of ultraviolet light (Seinfeld and Pandis, 2016).

The entire Woodside BAAMP dataset of hourly average NOx and O3 acquired from 2008 to 2015 was re- analysed. NOx is an expression of the total amount of both nitric oxide (NO) and NO2 in a gas, with the mass of NOx calculated by assuming all of the NO has been oxidised to NO2. Data capture for each pollutant, for each location, was an important consideration in the review. The results confirmed that typically NO2 exists in levels well below the relevant NEPM standard of 120 ppb. The monitoring results showed that O3 is a higher risk air pollutant for the Burrup Peninsula based on comparisons with the corresponding NEPM standard of 100 ppb.

The monitoring results showed higher O3 concentrations in Dampier and Karratha in comparison with NO2. An interpretation is NOx, assumed to be emitted primarily by Woodside sources, was dispersed to lower concentrations by the time it reached the townships of Dampier and Karratha. Therefore, there was less NOx in the townships to destroy the O3 that built up to higher concentrations there.

The locations of the Bureau of Meteorology (BoM) and Woodside air quality monitoring stations (Burrup Road, Dampier and Karratha), and the main, existing industrial facilities, are shown in Figure 3-3.

Final 14 Air Quality Impact Assessment

Figure 3-3: Locations of Meteorological and Air Quality Monitoring Stations

Final 15 Air Quality Impact Assessment

BAAMP data capture for NO2 and O3 for the three monitoring stations over 2009-2015 is set out in the following tables. Years for which no measurements occurred are indicated by ‘ND’ (No Data).

Table 3-1: Karratha Air Quality Monitoring – Data Capture NO2 and O3

Substance 2009 2010 2011 2012 2013 2014 2015

NO2 94% 95% 94% 97% 96% 93% 97%

O3 96% 95% 94% 94% 95% 92% 95%

Table 3-2: Dampier Air Quality Monitoring Results – Data Capture NO2 and O3

Substance 2009 2010 2011 2012 2013 2014 2015

NO2 92% 90% 92% 89% 93% 91% 94%

O3 30% 90% 96% 95% 95% 92% 96%

Table 3-3: Burrup Road Air Quality Monitoring Results – Data Capture NO2 and O3

Substance 2009 2010 2011 2012 2013 2014 2015

NO2 84% 94% 86% 90% 96% 94% 92%

O3 53% ND ND ND ND ND ND

Statistical summaries of the BAAMP results determined from hourly average NO2 concentrations for the three monitoring locations are illustrated in Figure 3-4 (Karratha), Figure 3-5 (Dampier), and Figure 3-6 (Burrup). The statistics determined from the hourly averages are: maximum, 99.9th percentile, etc., down to the median and annual averages. Inspection of the maximum hourly average and annual average NO2 concentrations (ppb) for the years shown in the three figures demonstrate clearly that there have been no exceedances of any NO2 standards over the monitoring period of several years. This includes 2014 when the Pluto LNG Plant (PLP) had ramped up to full production, and the Karratha Gas Plant (KGP) at the NWS Project was operating to capacity.

The NEPM (Ambient Air Quality) maximum hourly average NO2 standard is 120 ppb, and the annual average standard is 30 ppb.

Final 16 Air Quality Impact Assessment

Figure 3-4: Woodside Air Quality Monitoring Results 2009-2015: Karratha NO2

Figure 3-5: Woodside Air Quality Monitoring Results 2009-2015: Dampier NO2

Final 17 Air Quality Impact Assessment

Figure 3-6: Woodside Air Quality Monitoring Results 2009-2015: Burrup NO2

Statistical summaries of results for hourly average O3 concentrations are shown for the two monitoring locations where data capture was sufficient: Karratha (Figure 3-7) and Dampier (Figure 3-8). The corresponding NEPM standard (maximum hourly average, 100 ppb) was not exceeded in any hour measured over 2009-2015.

Figure 3-7: Woodside Air Quality Monitoring Results 2009-2015: Karratha O3

Final 18 Air Quality Impact Assessment

Figure 3-8: Woodside Air Quality Monitoring Results 2009-2015: Dampier O3

3.5 Hydrocarbons

Monitoring of hydrocarbons or Volatile Organic Compounds (VOCs) undertaken by Woodside over 2009-2015 showed that emissions of Benzene, Toluene and Xylenes (BTX), as an indicator of all VOCs, were insignificant where measurements were obtained in Dampier and Karratha. For most of the time, BTX concentrations were nil at those locations. From a review of hydrocarbon emissions and monitoring data, it was concluded that formaldehyde would have low concentrations with approximately the same low risk of air quality impact as benzene.

3.6 Sulfur Dioxide

A review of SO2 monitoring results on Burrup Peninsula was undertaken by Air Assessments (2010b). Conservative assumptions were applied to several fixed industrial emissions sources, noting very low sulfur-in- fuel concentrations. For this reason, estimates for exhaust SO2 for most sources are at or near the limit of detection, thus a reasonable estimate for an annual average would be 0.1 ppb. Maximum hourly average concentrations would not be expected to exceed 10 ppb for most locations away from engine exhausts on ships, the most significant source in the region.

3.7 Deposition Fluxes of Nitrogen and Sulfur

The deposition of air pollutants containing nitrogen and sulfur such as NH3, NO2 and SO2, is of interest to potential future studies investigating effects on land surfaces. On the Burrup Peninsula, Gillett (2008) determined total deposition flux of nitrogen and sulfur at a number of measurement sites in 2004/2005 and 2007/2008 by calculating the wet and dry deposition of several nitrogen and sulfur species in the gas and aqueous (rainwater) phases, including NH3, NO2, SO2, and nitric acid. The results for total N and S dry deposition from Gillett (2008) are illustrated in detail in Figure 3-9 (2004-2005 measurements), and Figure 3-10 (2007-2008 measurements). These CSIRO monitoring locations are shown plotted in Gillett (2008) and Gillett et al (2012).

Final 19 Air Quality Impact Assessment

Figure 3-9: Total N & S Dry Deposition Flux 2004-2005 (Gillett, 2008)

Figure 3-10: Total N & S Dry Deposition Flux 2007-2008 (Gillett, 2008)

Final 20 Air Quality Impact Assessment

From inspection of the results shown in Figure 3-9 and Figure 3-10, a conservative (high) estimate for total dry deposition for all gaseous pollutants was approximately 20 meq/m2/year, up to 2008. Note the Karratha township experiences the highest NH3 deposition; this may be due to emissions associated with handling or management of beef cattle (Shen et al., 2016 and Hacker et al., 2016), and/or the use of fertilisers (Gillett, 2008). The results for Dolphin Island and North Burrup indicate the background dry NH3 deposition is approximately 4 meq/m2/year.

A recent summary of deposition data provided by Woodside (2019) indicates the total nitrogen and sulfur deposition fluxes have changed little over two decades, even when the Woodside Pluto LNG Development and Karratha Gas Plants were operating at or near capacity in 2014. A summary of results for nitrogen and sulfur deposition on Burrup Peninsula is provided in the following points (Woodside, 2019): · Total N and S deposition, 2004-2005 and 2007-2008, 20–32 meq/m2/year. · Total N deposition, 2008-2009, 18–33 meq/m2/year. · Total N deposition, 2012-2014, 17–29 meq/m2/year.

Deposition results were reported by Yara Pilbara Nitrates (YPN) and Strategen Environmental (YPN, 2017). A summary of their 2013-2016 results for dry deposition of HNO3, NH3, NO2 and SO2; i.e., the same components shown in the previous figures, are shown in Figure 3-11. These show a similar range for total dry deposition of around 20-30 meq/m2/year,even with the effects of most of the current industry on the Burrup Peninsula. In summary, the background total (dry and wet) nitrogen and sulfur deposition for Burrup Peninsula ranges between approximately 20–30 meq/m2/year, with a conservative (high) estimate of 30 meq/m2/year (by inspection of these results).

Figure 3-11: Summary of Total N & S Dry Deposition Flux 2003-2016 (YPN, 2017)

Final 21 Air Quality Impact Assessment

4. Emissions Sources and Estimates

4.1 Overview

The principal emissions from the PU Project will arise from combustion of natural gas and the production of urea. The most significant products of gas combustion include: carbon dioxide (CO2), NOx, carbon monoxide (CO) and unburnt hydrocarbons (VOCs). Significant air pollutants associated with urea production are: NOx, PM10, PM2.5, NH3, formaldehyde (CH2O), and methanol.

A broad-level risk assessment was conducted to confirm the key air pollutants and sources specifically for the PU Project. The purpose was to determine the relative risk of air pollutants and emission sources in proximity to the PU Project, with a focus on the Burrup Peninsula and the surrounding region. Previous air assessments and other relevant publicly available information were reviewed, as a part of validation of the existing air quality environment and model inputs. The outcomes of this risk assessment identified precisely what facilities should be included in the modelling and what substances should be modelled.

Emission inventories were developed in consultation with PCF, based on reasonable and conservative emission estimates, consideration of available datasets, design data, and monitoring data for the PU Project. Representative third-party emissions were based on consideration of publicly available literature and input following consultation with external parties.

4.2 Risk Assessment

A risk assessment was undertaken based on a broad survey of Burrup Peninsula air quality studies, emission inventories and other information, to determine key air pollutants and air emission sources. Emissions estimates for the PU Project were then compared to a consistent set of air quality standards in the context of regional air quality. Overall, the risk assessment determined that the key substances in relation to emissions from the PU Project were: PM10, PM2.5, NH3 and NO2, (and O3 by association with NO2), with lower risks associated with emissions of SO2, methanol, and the VOCs–formaldehyde was identified as being representative of the highest risk VOCs.

The substances selected for assessment for the PU Project were those substances identified in the risk assessment, with a stronger focus on the higher risk substances PM10, PM2.5, NH3, NO2 and O3. Model results were obtained for SO2, methanol and formaldehyde also.

In relation to VOCs, monitoring undertaken by Woodside over 2009-2015 showed that emissions of BTX, as an indicator of VOCs, had insignificant air quality effects at the sensitive receptor locations of Dampier and Karratha. For most of the time, BTX concentrations were nil at those locations. It was concluded that formaldehyde would have low concentrations that were approximately the same as benzene. However, estimates for total VOC emissions were included in the modelling as a part of the input for the photochemical modelling.

Regional (beyond the Burrup Peninsula) emission sources were excluded from the air quality assessment as previous modelling studies demonstrated that, while there may be some transfer of air pollutants, these would be minimal, given the distance. The Air Assessments (2012) results clearly show that air quality effects on the Burrup Peninsula are primarily due to sources on the Burrup Peninsula. In any case, the air quality effects from smaller or lower risk sources were accounted for to some extent by the inclusion of background air pollutant concentrations in the modelling. The lower risk sources fell into the following classes: · Too small as emitters by mass. · Too distant for the dispersed pollutants to make a significant contribution to ambient levels around the Burrup Peninsula; e.g. beyond approximately 50 km from Dampier and Karratha. · Substances emitted not associated with air quality effects caused by emissions from the PU Project processing facilities.

Final 22 Air Quality Impact Assessment

The risk assessment also demonstrated that emissions from regional shipping have the potential to make a significant contribution to ambient NOx levels and need to be considered in the modelling.

Based on the findings of the risk assessment, 94 existing air pollutant point sources (stack) on the Burrup Peninsula were identified to be included in the modelling. A summary of these point sources, with total NOx emissions (g/s), is presented in Table 4-1. Emission source locations are provided in Figure 4-1.

Table 4-1: Summary of Current Air Emissions Sources Considered in the Modelling Assessment

Number of Total NOx Emission Industrial Facility Emission Sources Rate (g/s)

Karratha Gas Plant (Woodside) 44 281

Pluto LNG Plant (Woodside) 11 34.1

Yara Technical Ammonium Nitrate and Liquid Ammonium Plant 5 30.3

Pilbara Iron Yurralyi Maya Power Station 5 28.2

Santos Devil Creek Power Station 7 4.5

ATCO Karratha Power Station 2 12.0

EDL West Kimberley Power Plant 3 1.2

All shipping berths on the Burrup Peninsula 13 26.0

All the main shipping berths at Cape Lambert 5 10.0

The estimates for numbers of emissions sources and total NOx emissions for each facility listed in Table 4-1 were obtained from a variety of sources. Data were provided direct from Woodside for the KGP and Pluto LNG Plant. For the non-Woodside sources a variety of references were used to estimate emissions including EPA assessment reports for some of the larger facilities, the most recent annual National Pollutant Inventory reports, and vertical imagery was used to inspect each of the sites. More details are provided in Section 4.4.

Final 23 Air Quality Impact Assessment

Figure 4-1 Locations of Modelled Emissions Sources

Final 24 Air Quality Impact Assessment

4.3 Model Scenarios

Eight air emissions scenarios were tested by modelling to assess the PU Project. These scenarios are detailed in Table 4-2. Further details about specific sources for modelling are set out in the following sub-sections. The modelled air emissions scenarios are based on predicting cumulative impacts on the Burrup Peninsula arising from current and future industrial activity and consider various development options.

Table 4-2: Air Emissions Scenarios for Assessment

Scenario Description and Emission Sources

(1) Perdaman in isolation The PNO scenario represents the PU Project in isolation as the sole emission source, operating under (Normal operations conditions) normal conditions. (PNO)

(2) Perdaman in isolation The PUC scenario represents the PU Project in isolation as the sole emission source, operating under (upset conditions) the upset condition of flaring, over a period of 1-2 weeks annually. (PUC)

(3) Perdaman normal The Total Ammonia scenario represents the implementation of the PU Project operating under normal operations plus Yara Pilbara conditions alongside the existing Yara Pilbara facilities on the Burrup Peninsula. sources (NH3 only) Air emissions estimates for the following currently operating facilities are included: (Total Ammonia) · Yara Technical Ammonium Nitrate and Liquid Ammonium Plant The Total Ammonia scenario represents modelling of NH3 emissions based on a ‘low-NH3’ emissions scenario representing operation of the plant with mitigation (e.g. acid scrubbing) in place from day one. It should be noted that the modelled emissions are considered to be representative of the second year of operations and onwards, as the first year is likely to include testing and commissioning with reduced operational time and assumed lower emissions.

(4) Existing air emissions The Baseline scenario represents all current relevant air pollutant sources. (Baseline) Baseline represents the existing air emissions scenario mostly applicable to the Burrup Strategic Industrial Area (BSIA) and the region to use as a baseline for assessment. Air emissions estimates for the following currently operating facilities are included: · KGP · PLP · Yara Technical Ammonium Nitrate and Liquid Ammonium Plant · Pilbara Iron Yurralyi Maya Power Station · Santos Devil Creek Power Station · ATCO Karratha Power Station · EDL West Kimberley Power Plant · All shipping berths on the Burrup Peninsula · All shipping berths at Cape Lambert Baseline represents the current and near-term operating scenario and could be described as a ‘near- term most likely’ case.

(5) Baseline condition including The BPNO scenario reflects the implementation of the PU Project under normal operating conditions Perdaman (Normal operations and includes cumulative impacts from current facilities on the Burrup Peninsula. conditions) This scenario includes emissions from: (BPNO) · All the currently operating facilities listed above under the Baseline scenario · The implementation of the PU Project under normal operating conditions · Future development on the Burrup Peninsula is not considered. It is considered to be a ‘most likely’ and ‘best case’ for future ambient air quality on the Burrup Peninsula.

Final 25 Air Quality Impact Assessment

Scenario Description and Emission Sources

(6) Baseline condition including The BPUC scenario reflects the implementation of the PU Project under upset operating conditions Perdaman (upset conditions) and includes cumulative impacts from current facilities on the Burrup Peninsula. (BPUC) This scenario includes emissions from: · All the currently operating facilities listed above under the Baseline scenario · The implementation of the PU Project under upset operating conditions · Future development on the Burrup Peninsula is not considered.

(7) Baseline condition including The FPNO scenario reflects the implementation of the PU Project and includes cumulative impacts Perdaman and other proposed from current and likely future facilities on the Burrup Peninsula. projects This scenario includes emissions from: (FPNO) · All the currently operating facilities listed above under the CB scenario · The implementation of the PU Project under normal operating conditions · Future development on the Burrup Peninsula represented by the Methanol plant. It is considered to be the ‘most likely’ and ‘worst case’ for future ambient air quality on the Burrup Peninsula.

(8) PNO particulate urea For determination of particulate urea deposition due to emissions from the PU Project granulator vents,

emissions as PM10 assuming urea is the sole particulate species from those vents (100% of PM10 emissions). While emitted urea dust may decompose after emissions, as a worst-case scenario, no degradation was assumed. Also, as urea is basic, and not an acid-forming nitrate, cumulative modelling with ammonium nitrate emissions, e.g. from the Yara plants, was not included.

4.4 Existing Emission Sources

4.4.1 Karratha Gas Plant

The Karratha Gas Plant (KGP) was originally commissioned in 1984 with feed gas and fluid sources from the North Rankin platform. The KGP has undergone a number of expansions and additional facilities have been installed since it was first commissioned. At present the existing NWS Project processes natural gas and associated fluids from NWSJV field resources to produce up to 18.5 million tonnes per annum (Mtpa) of Liquefied Natural Gas at the KGP.

The existing KGP air emission sources comprise: · Four domestic gas (Domgas) GTCs. · Trains 1, 2 and 3 - each consisting of five GTCs, with one GTC exhaust per train with integrated Acid Gas Removal Unit (AGRU) CO2 vent stack system. · Trains 4 and 5 – each consisting of two GTCs, with one machine each including two WHRU exhaust stacks.

· 10 power generation gas turbines, with two providing integrated AGRU CO2 vent stack systems for LNG Trains 4 and 5.

Air emission parameters for the KGP sources are listed in Table 4-3 (data provided by Woodside).

Final 26 Air Quality Impact Assessment

Table 4-3: NWS Karratha Gas Plant Air Emissions Parameters

Stack Stack Exit Temp. PM10 NOx SO2 Emissions Source Height Diameter Velocity VOC (g/s) (K) (g/s) (g/s) (g/s) (m) (m) (m/s)

Domgas GTC 1 24 1.95 42.3 815 0.01 3.81 0.12 0.01

Domgas GTC 2 24 2.8 43.4 764 0.01 12.02 0.25 0.01

Domgas GTC 3 24 1.95 42.3 815 0.01 3.81 0.12 0.01

Domgas GTC 4 24 2.8 43.4 764 0.01 12.02 0.25 0.01

TRAIN 1 – GTC 1 40 3.88 19.5 777 0.01 10.15 0.27 0.01

TRAIN 1 – GTC 2 40 3.88 19.5 782 0.01 9.68 0.27 0.01

TRAIN 1 – GTC 3 40 3.6 22.7 767 0.01 9.81 0.27 0.01

TRAIN 1 – GTC 4 40 3.6 21.7 771 0.01 9.19 0.27 13.53

TRAIN 1 – GTC 5 40 2.72 18.9 795 0.01 3.55 0.12 0.01

TRAIN 2 – GTC 1 40 3.88 19.5 777 0.01 10.15 0.27 0.01

TRAIN 2 – GTC 2 40 3.88 19.5 782 0.01 9.68 0.27 0.01

TRAIN 2 – GTC 3 40 3.6 22.7 767 0.01 9.81 0.27 0.01

TRAIN 2 – GTC 4 40 3.6 21.7 771 0.01 9.19 0.27 13.53

TRAIN 2 – GTC 5 40 2.72 18.9 795 0.01 3.55 0.12 0.01

TRAIN 3 – GTC 1 40 3.88 19.5 777 0.01 10.15 0.27 0.01

TRAIN 3 – GTC 2 40 3.88 19.5 782 0.01 9.68 0.27 0.01

TRAIN 3 – GTC 3 40 3.6 22.7 767 0.01 9.81 0.27 0.01

TRAIN 3 – GTC 4 40 3.6 21.7 771 0.01 9.19 0.27 13.53

TRAIN 3 – GTC 5 40 2.72 18.9 795 0.01 3.55 0.12 0.01

TRAIN 4 – GTC 2 40 6 23.8 811 0.01 5.79 0.64 0.01

TRAIN 4 – GTC 1 WHRU1 40 2.9 50.9 588 0.01 3.13 0.29 0.01

TRAIN 4 – GTC 1 WHRU2 40 2.9 50.9 521 0.01 3.13 0.29 0.01

TRAIN 5 – GTC 2 40 6.01 23.7 811 0.01 7.18 0.64 0.01

TRAIN 5 – GTC 2 WHRU 1 40 2.9 50.9 523 0.01 3.11 0.29 0.01

TRAIN 5 – GTC 1 WHRU 2 40 2.9 50.9 483 0.01 3.11 0.29 0.01

Stabiliser 2 Furnace Stack 33 1.46 39.2 699 0.01 2.56 0.01 0.01

Stabiliser 4 Furnace Stack 33 1.46 39.2 668 0.01 2.17 0.01 0.01

Stabiliser 5 Furnace Stack 33 1.46 39.2 659 0.01 2.23 0.01 0.01

Stabiliser 6 Furnace Stack 33 1.46 39.2 630 0.01 1.98 0.01 0.01

Power Generation GTG1 40 3.95 20.4 681 0.01 11.58 0.24 0.01

Power Generation GTG2 40 3.95 21.5 681 0.01 12.21 0.24 0.01

Power Generation GTG3 40 3.95 20.4 675 0.01 8.63 0.24 0.01

Final 27 Air Quality Impact Assessment

Stack Stack Exit Temp. PM10 NOx SO2 Emissions Source Height Diameter Velocity VOC (g/s) (K) (g/s) (g/s) (g/s) (m) (m) (m/s)

Power Generation GTG4 40 3.95 21.5 681 0.01 12.21 0.24 0.01

Power Generation GTG5 40 3.95 20.4 675 0.01 8.63 0.24 0.01

Power Generation GTG6 40 3.95 20.4 675 0.02 8.63 0.24 40.6

Power Generation 7 40 3.57 22.2 751 0.01 3 0.22 0.01

Power Generation 8 40 3.57 17.7 751 0.01 2.66 0.22 40.6

Power Generation 9 40 3.57 34.6 751 0.01 4.45 0.22 0.01

Power Generation 10 40 3.57 31.3 745 0.01 3.64 0.22 0.01

Domgas-E Flare 128.5* 1.03 20 1273 0.05 0.28 0.001 0.58

LNG Emergency Flare 145.3* 6.526 20 1273 1.95 11.32 0.044 23.42

LNG-SL Flare 56.9* 0.55 20 1273 0.01 0.08 0.0003 0.17

LPG-SL Flare 56.5* 0.426 20 1273 0.01 0.05 0.0002 0.10

Operations Flare 46.8* 1.456 20 1273 0.10 0.56 0.002 1.17 *Calculated effective stack height for flare sources; USEPA (1992); USEPA (1995).

Flare emissions are represented conservatively with elevated rate applied for KGP Emergency Flare as a constant source in the model to reflect potential for frequent intermittent operation across the KGP and Pluto LNG Plant. Credible baseload flaring is assumed for other flare points (Woodside).

4.4.2 Woodside Pluto Onshore LNG Plant

The Pluto gas field was discovered in April 2005 and is located on the North West Shelf of WA, approximately 190 km north-west of Dampier. The associated gas processing plant is located on the Burrup Peninsula, approximately 6 km from Dampier.

The Pluto LNG Development was approved by the EPA and Commonwealth Department of Environment and Energy (DoEE) following public environment review of the PU Project in 2006. The original proposal included the construction, commissioning and operation of the Pluto LNG Development with two LNG processing trains. However only one train was built, commissioned and operated.

The Woodside PLP air emissions parameters are listed in Table 4-4 (data provided by Woodside).

Final 28 Air Quality Impact Assessment

Table 4-4: Pluto Onshore LNG Plant Air Emissions Parameters

Stack Stack Exit Temp. PM10 NOx SO2 VOC Emissions Source Height Diameter Velocity (K) (g/s) (g/s) (g/s) (g/s) (m) (m) (m/s)

PLP Train 1 – GTC 1 WHRU 1 40 2.90 39.2 531.2 0.01 5.63 0.37 0.01

PLP Train 1 GTC 1 WHRU 2 40 2.90 41.2 527.2 0.01 5.10 0.38 0.01

PLP Train 1 – GTC 2 40 6.01 28.0 824.2 0.01 10.20 0.37 0.01

PLP GTG 1 40 3.11 28.0 868.2 0.01 3.27 0.25 0.01

PLP GTG 2 40 3.86 23.0 874.2 0.01 3.36 0.24 0.01

PLP GTG 3 40 2.80 23.8 879.2 0.01 3.22 0.16 0.01

PLP GTG 4 40 2.80 22.0 883.2 0.01 1.82 0.33 0.01

PLP Train 1 - Regenerative Thermal 40 2.80 17.7 394.2 0.01 0.08 0.42 0.01 Oxidiser

Flare Cold Dry 139.5* 1.34 20.0 1273 0.08 0.49 0.002 1.010

Flare Warm Wet 139.5* 1.34 20.0 1273 0.08 0.49 0.002 1.013

Storage and Loading Flare 64.3* 1.28 20.0 1273 0.08 0.45 0.002 0.923 *Calculated effective stack height for flare sources; USEPA (1992); USEPA (1995).

Flare emissions are represented conservatively with elevated rate applied for KGP Emergency Flare as a constant source in the model to reflect potential for frequent intermittent operation across the KGP and Pluto LNG Plant. Credible baseload flaring is assumed for other flare points (Woodside).

4.4.3 Other Air Emissions Sources

The risk assessment determined that point source (stack) emissions of NOx, VOCs and other substances from the following facilities have the potential to make a significant contribution to the ground level concentrations and therefore needed to be considered in any air quality assessment: · Yara fertilisers and nitrate facilities · Pilbara Iron Yurralyi Maya Power Station · Santos Devil Creek Power Station · ATCO Karratha Power Station · West Kimberley Power Plant

A variety of methods was used to identify and locate the point sources (stacks) of these facilities, and determine stack and air emissions parameters for modelling, including: review and analysis of annual NPI reports, review of previous EPA assessment reports, and review of EPA air discharge licences. A stand-alone air quality assessment report could have been written for each of these non-Perdaman facilities – for brevity, the following sections are limited to setting out the final emissions parameters and estimates used as inputs for modelling.

The Yara Pilbara Fertilisers and Yara Pilbara Nitrates TAN air emissions parameters are listed in Table 4-5, based on a review of previous assessment reports and EPA licences: GWA (2015); GWA (2018); Burrup Nitrates (2009a); Burrup Nitrates (2009b); EPA (2011); and EPA (2001). Emissions estimates for VOCs from the Yara plants were nil (these references).

Final 29 Air Quality Impact Assessment

Table 4-5: Yara Pilbara Fertiliser and Yara Pilbara Nitrates TAN Air Emissions Parameters

Stack Stack Exit

Emissions Source Height (m) Diameter Velocity Temp. (K) PM10 (g/s) NOx (g/s) SO2 (g/s) NH3 (g/s) (m) (m/s)

TAN Plant Stack - Prilling 70 1.9 11.9 309 0.8 0 0 0.639 Tower

TAN Nitric Acid Plant 54 1.4 27.5 423 0 4.2 0 0.022 Stack

TAN ‘offsite’ power 30 2.6 16.9 450 0.058 2.1 0 0 generation

Fertiliser Reformer 35 3.5 15.0 413 0.91 17.1 0.23 0

Fertiliser Boiler 30 3 4.1 450 0.36 6.9 0.13 0

The Yurralyi Maya Power Station, owned and operated by Hamersley Iron Pty Ltd, is located approximately 17 km south of the Burrup Hub site. Key air emissions sources of the Yurralyi Maya Power Station are the gas turbines; air emissions parameters are listed in Table 4-6.

Table 4-6: Yurralyi Maya Power Station Emissions Data

Stack Exit Stack Emissions Source Diameter Velocity Temp. (K) PM10 (g/s) NOx (g/s) SO2 (g/s) VOC (g/s) Height (m) (m) (m/s)

GTG 1 40 3.57 25.7 722 1 5.63 4.00 0.04

GTG 2 40 3.57 25.7 722 1 5.63 4.00 0.04

GTG 3 40 3.57 25.7 722 1 5.63 4.00 0.04

GTG 4 40 3.57 25.7 722 1 5.63 4.00 0.04

GTG 5 40 3.57 25.7 722 1 5.63 4.00 0.04

The Devil Creek Gas Plant, operated by Santos (formerly Quadrant Energy), is located 48 km south west of the Burrup hub site. The Devil Creek Gas Plant equipment identified as key air emission sources for the BHSM were: · two Solar Taurus 60 gas turbine generators of nominal 5000 kW capacity providing electrical power requirements. · two sales gas compressors power by Solar Taurus 60 gas turbines, fitted with waste heat recovery units; · waste gas incinerator. · and an elevated flare and ground flare.

The associated air emissions parameters are listed in Table 4-7.

Table 4-7: Devil Creek Gas Plant Air Emissions Parameters

Exit Stack Stack PM10 VOC Emissions Source Velocity Temp. (K) NOx (g/s) SO2 (g/s) Height (m) Diameter (m) (g/s) (g/s) (m/s)

GTG 1 13 1.6 23.5 783 0.004 0.75 0.0 0.005

Final 30 Air Quality Impact Assessment

Exit Stack Stack PM10 VOC Emissions Source Velocity Temp. (K) NOx (g/s) SO2 (g/s) Height (m) Diameter (m) (g/s) (g/s) (m/s)

GTG 2 13 1.6 23.5 783 0.004 0.75 0.0 0.005

GTC 1 13 1.6 16.0 633 0.004 0.75 0.0 0.005

GTC 2 13 1.6 16.0 633 0.004 0.75 0.0 0.005

Waste Gas Incinerator 21 1.8 14.0 1073 0.004 0.00 10.96 0.005

Elevated Flare 48 1.62 20.0 1273 0.004 0.77 0.0 0.005

Ground Flare 20 1.62 20.0 1273 0.004 0.77 0.0 0.005

The West Kimberley Power Station, operated by EDL Energy, is located approximately 25 km south-west of the Burrup Hub site. Air emissions parameters for the three gas turbines, are listed in Table 4-8.

Table 4-8: West Kimberley Power Project Emissions Data

Exit Stack Stack PM10 VOC Emissions Source Velocity Temp. (K) NOx (g/s) SO2 (g/s) Height (m) Diameter (m) (g/s) (g/s) (m/s)

GTG 1 10 1.2 26.5 700 0.002 0.385 0.0006 0.0025

GTG 2 10 1.2 26.5 700 0.002 0.385 0.0006 0.0025

GTG 3 10 1.2 26.5 700 0.002 0.385 0.0006 0.0025

The ATCO Karratha Power station is located 18 km south-east of the Burrup Hub site. Key air emissions sources identified were two LM6000 DP Sprint gas turbines; the air emissions parameters are listed in Table 4-9.

Table 4-9: Karratha Power Station Emissions Data

Exit Stack Stack PM10 VOC Emissions Source Velocity Temp. (K) NOx (g/s) SO2 (g/s) Height (m) Diameter (m) (g/s) (g/s) (m/s)

GTG 1 18.2 3.57 26.0 723 0.04 6 0.01 0.043

GTG 2 18.2 3.57 26.0 723 0.04 6 0.01 0.043

Emissions from shipping were modelled for all thirteen berths on the Burrup Peninsula, and five berths at Cape Lambert. A ship was assumed to be docked at all these berths with ancillary engines running continuously; i.e. 24 hours per day, every day of the year. The air emissions parameters assigned for each of the total of eighteen berth locations are listed in Table 4-10 (these are emissions parameters for a single berth; examples are the NOx sub-totals: Burrup Peninsula berths 26 g/s; Cape Lambert berths 10 g/s).

Table 4-10: Air Emissions Data for Shipping

Stack Stack Temperature VOC Diameter EV (m/s) PM10 PM2.5 NOx (g/s) CO (g/s) SO2 (g/s) Height (m) (K) (g/s) (m)

35 0.5 11.9 673 0.25 0.23 2.0 0.33 2.0 0.12

Final 31 Air Quality Impact Assessment

4.5 Future Emission Sources

4.5.1 Overview

Modelling conducted for the “future” scenarios included emissions from various combinations of the following sources: · The PU Project · Woodside Pluto LNG Development: existing Train 1 and proposed Train 2 expansion (preliminary design 2019) · Woodside KGP with improvement assumptions · A proposed new Methanol Plant.

4.5.2 Perdaman Urea Project The PU Project is proposed to be located within the BSIA with a production capacity of approximately 2 Mtpa on Sites C and F within the BSIA, with natural gas for the PU Project to be sourced from a nearby domestic gas plant (Section 1.1). The air emissions sources and air emissions parameters for modelling were identified and set out from an analysis of engineering and other data provided by Cardno and Perdaman over June-July 2019. The key sources were a fired heater, two gas turbines, two urea train absorber vents, and two urea granulator stacks; further details may be found in Cardno (2019).

Key air emissions parameters for modelling for the PU Project are set out in the following tables: common parameters (Table 4-11); normal operations (Table 4-12), and upset conditions (Table 4-13).

Upset conditions for the plant or “turndown mode” is defined as operations at approximately 60% of the normal rate. The normal operating parameters for the plant are typically near 100% of design. The ammonia plant is designed to precisely control the gas inputs to the various process units, within varying operating conditions. These include amounts of natural gas, steam, oxygen and nitrogen inputs to the process. Effectively, most emissions are proportional to process output. This results in reduced mass emission rates for each substance during upset conditions as the various process units are operated at reduced output. However, it is noted the overall energy intensity and emissions per tonne of ammonia are higher under upset conditions than for normal operations.

Table 4-11: PU Project Air Emissions Sources and Parameters – Common Data

MGA94 MGA94 Northing Stack height Stack Exhaust temp. Emissions Source Easting (m) (m) (m) diameter (m) (oC)

Fired Heater H201 476,637 7,718,899 75 2.7 120

Gas Turbine Generator 1 476,748 7,718,808 30.5 3.4 85

Gas Turbine Generator 2 476,748 7,718,790 30.5 3.4 85

Urea Train 1 Absorber vent 476,335 7,718,972 40 0.2 43

Urea Train 2 Absorber vent 476,335 7,718,862 40 0.2 43

Urea Train 1 Granulator stack 476,310 7,718,978 40 4.2 42

Urea Train 2 Granulator stack 476,310 7,718,868 40 4.2 42

Final 32 Air Quality Impact Assessment

Table 4-12: PU Project Air Emissions Sources and Air Emissions Estimates – Normal Operations

Emissions Exit vel. NOx PM10 NH3 SO2 Methanol CH2O CO VOC (g/s) Source (m/s) (g/s) (g/s) (g/s) (g/s) (g/s) (g/s) (g/s)

Fired Heater H201 16.6 6.68 0.13 0.01 0 0.04 0 0.003 2.73

Gas Turbine 21.0 2.49 0.21 0.005 0 0.07 0 0.0035 1.47 Generator 1

Gas Turbine 21.0 2.49 0.21 0.005 0 0.07 0 0.0035 1.47 Generator 2

Urea Train 1 15.8 0 0 0 1.8 0 0 0 0 Absorber vent

Urea Train 2 15.8 0 0 0 1.8 0 0 0 0 Absorber vent

Urea Train 1 20.1 0 5.43 0 4.26 0 0.003 0.003 0 Granulator stack

Urea Train 2 20.1 0 5.43 0 4.26 0 0.003 0.003 0 Granulator stack

Table 4-13: PU Project Air Emissions Sources and Air Emissions Estimates – Upset Conditions

Emissions Exit vel. NOx PM10 NH3 Methanol CH2O CO VOC (g/s) SO2 (g/s) Source (m/s) (g/s) (g/s) (g/s) (g/s) (g/s) (g/s)

Fired Heater H201 10.2 4.21 0.08 0.007 0 0.03 0 0.0021 1.75

Gas Turbine 14.7 1.74 0.14 0.0037 0 0.05 0 0.0026 1.03 Generator 1

Gas Turbine 14.7 1.74 0.14 0.0037 0 0.05 0 0.0026 1.03 Generator 2

Urea Train 1 9.9 0 0 0 1.08 0 0 0 0 Absorber vent

Urea Train 2 9.9 0 0 0 1.08 0 0 0 0 Absorber vent

Urea Train 1 13.9 0 6.62 0 5.11 0 0.0018 0.0018 0 Granulator stack

Urea Train 2 13.9 0 6.62 0 5.11 0 0.0018 0.0018 0 Granulator stack

Particulate urea deposition due to emissions from the PU Project granulator vents was determined with the assumption that urea was the only particulate species from those vents (100% of PM10 emissions). While emitted urea dust may decompose after emissions, as a worst-case scenario, no degradation was assumed. Also, as urea is basic, and not an acid-forming nitrate, cumulative modelling with ammonium nitrate emissions; i.e., from the Yara plants, was not included. A PM2.5/PM10 ratio of 30% for the urea dust particles was assumed for the assessment, for consistency with the GHD (2009) assessment for the proposed Collie Urea Project.

Final 33 Air Quality Impact Assessment

4.5.3 Woodside Pluto Development

Woodside, as operator of the Pluto LNG Development, is proposing the Pluto Expansion Project. This includes the construction and commissioning of a second LNG processing train, Pluto Train 2. The emissions inventory used as input for the modelling is provided in Table 4-14 (data provided by Woodside). Also, minor changes to Train 1 were included in the modelling.

Table 4-14: Pluto LNG Development – Train 2 Air Emissions Parameters

Stack Stack Exit Temp. VOC Emissions Source Height Diamete Velocity NOx (g/s) SO2 (g/s) (K) (g/s) (m) r (m) (m/s)

Train 1 – GTG 3 40.1 2.80 29.1 821 2.98 0.07 0.01

Train 1 – GTG 4 40.1 2.80 29.5 823 3.53 0.06 0.01

Train 2 – GTC 1 50.7 3.06 29.6 741 4.55 0.002 0.01

Train 2 – GTC 2 50.7 3.06 29.6 741 4.55 0.002 0.01

Train 2 – GTC 3 50.7 3.6 2.4 741 4.55 0.002 0.01

Train 2 – GTC 4 50.7 3.06 29.6 584 4.55 0.002 0.01

Train 2 – GTC 5 50.7 3.6 2.4 741 4.55 0.002 0.01

Train 2 – GTC 6 50.7 3.06 29.6 584 4.55 0.002 0.01

PLP GTG 5 30.0 5.7 38.3 787 4.88 0.003 0.01

PLP Train 2 - AGRU Thermal Oxidiser 16.0 0.84 13.2 962 0.69 0.141 0.01

PLP Train 2 - NRU Thermal Oxidiser 30.5 1.07 31.0 700 0.70 0.040 0.01

4.5.4 Woodside KGP Development

The NWS Project Extension Proposal includes a reduction in NOx emissions. The complete emissions inventory including modifications for input to the modelling is provided in Table 4-15 (data provided by Woodside).

Table 4-15: Karratha Gas Plant – NWS Project Extension Proposal

Stack Stack Exit Temp. PM10 VOC Emissions Source Height Diameter Velocity NOx (g/s) SO2 (g/s) (K) (g/s) (g/s) (m) (m) (m/s)

Domgas GTC 1 24.0 1.0 42.3 815 0.01 3.81 0.12 0.01

Domgas GTC 2 24.0 1.4 43.4 764 0.01 4.47 0.25 0.01

Domgas GTC 3 24.0 1.0 42.3 815 0.01 3.81 0.12 0.01

Domgas GTC 4 24.0 1.4 43.4 764 0.01 4.47 0.25 0.01

TRAIN 1 – GTC 1 40.0 1.9 23.1 764 0.01 4.47 0.27 0.01

TRAIN 1 – GTC 2 40.0 1.9 23.1 764 0.01 4.47 0.27 0.01

TRAIN 1 – GTC 3 40.0 1.8 26.9 764 0.01 4.47 0.27 0.01

TRAIN 1 – GTC 4 40.0 1.8 26.9 764 0.01 4.47 0.27 0.01

TRAIN 1 – GTC 5 40.0 1.4 18.9 795 0.01 3.55 0.12 0.01

Final 34 Air Quality Impact Assessment

Stack Stack Exit Temp. PM10 VOC Emissions Source Height Diameter Velocity NOx (g/s) SO2 (g/s) (K) (g/s) (g/s) (m) (m) (m/s)

TRAIN 2 – GTC 1 40.0 1.9 23.1 764 0.01 4.47 0.27 0.01

TRAIN 2 – GTC 2 40.0 1.9 23.1 764 0.01 4.47 0.27 0.01

TRAIN 2 – GTC 3 40.0 1.8 26.9 764 0.01 4.47 0.27 0.01

TRAIN 2 – GTC 4 40.0 1.8 26.9 764 0.01 4.47 0.27 0.01

TRAIN 2 – GTC 5 40.0 1.4 18.9 795 0.01 3.55 0.12 0.01

TRAIN 3 – GTC 1 40.0 1.9 23.1 764 0.01 4.47 0.27 0.01

TRAIN 3 – GTC 2 40.0 1.9 23.1 764 0.01 4.47 0.27 0.01

TRAIN 3 – GTC 3 40.0 1.8 26.9 764 0.01 4.47 0.27 0.01

TRAIN 3 – GTC 4 40.0 1.8 26.9 764 0.01 4.47 0.27 0.01

TRAIN 3 – GTC 5 40.0 1.4 18.9 795 0.01 3.55 0.12 0.01

TRAIN 4 – GTC 2 40.1 3.0 23.8 811 0.01 5.79 0.64 0.01

TRAIN 4 – GTC 1 WHRU1 40.1 1.5 50.9 588 0.01 3.13 0.29 0.01

TRAIN 4 – GTC 1 WHRU2 40.1 1.5 50.9 521 0.01 3.13 0.29 0.01

TRAIN 5 – GTC 2 40.1 3.0 23.7 811 0.01 7.18 0.64 0.01

TRAIN 5 – GTC 1 WHRU 1 40.1 1.5 50.9 523 0.01 3.11 0.29 0.01

TRAIN 5 – GTC 1 WHRU 2 40.1 1.5 50.9 483 0.01 3.11 0.29 0.01

Stabiliser 2 Furnace Stack 33.0 0.7 39.2 699 0.01 2.56 0.01 0.01

Stabiliser 4 Furnace Stack 33.0 0.7 39.2 668 0.01 2.17 0.01 0.01

Stabiliser 5 Furnace Stack 33.0 0.7 39.2 659 0.01 2.23 0.01 0.01

Stabiliser 6 Furnace Stack 32.6 0.7 39.2 630 0.01 1.98 0.01 0.01

Power Generation GTG1 40.0 2.0 17.1 814 0.01 2.01 0.24 0.01

Power Generation GTG2 40.0 2.0 17.1 814 0.01 2.01 0.24 0.01

Power Generation GTG3 40.0 2.0 17.1 814 0.01 2.01 0.24 0.01

Power Generation GTG4 40.0 2.0 17.1 814 0.01 2.01 0.24 0.01

Power Generation GTG5 40.0 2.0 17.1 814 0.01 2.01 0.24 0.01

Power Generation GTG6 40.0 2.0 17.1 814 0.02 2.01 0.24 40.61

Power Generation 7 40.0 1.8 22.2 751 0.01 3.00 0.22 0.01

Power Generation 8 40.0 1.8 17.7 751 0.01 2.66 0.22 40.60

Power Generation 9 40.0 1.8 34.6 751 0.01 4.45 0.22 0.01

Power Generation 10 40.0 1.8 31.3 745 0.01 3.64 0.22 0.01

Domgas-E Flare 128.5 0.5 20.0 1273 0.05 0.28 0.00 0.58

LNG Emergency Flare 145.3 3.3 20.0 1273 1.95 11.32 0.04 23.42 (representative source)

Final 35 Air Quality Impact Assessment

Stack Stack Exit Temp. PM10 VOC Emissions Source Height Diameter Velocity NOx (g/s) SO2 (g/s) (K) (g/s) (g/s) (m) (m) (m/s)

LNG-SL Flare 56.9 0.3 20.0 1273 0.01 0.08 0.00 0.17

LPG-SL Flare 56.5 0.2 20.0 1273 0.01 0.05 0.00 0.10

Operations Flare 46.8 0.7 20.0 1273 0.10 0.56 0.00 1.17

4.5.5 Proposed Methanol Plant

The future air emissions scenario for modelling includes one other representative facility located within the BSIA, near the PU Project, a proposed methanol plant with production capacity of approximately 5,000 tonnes per day. Air emissions parameters used in the modelling for the proposed methanol plant are listed in Table 4-16.

Table 4-16: Air Emissions Data for Methanol Proposal

Stack Stack Exit Temp. NOx SO2 VOC Emissions Source Height Diameter Velocity (K) (g/s) (g/s) (g/s) (m) (m) (m/s)

Flue Gas Stack 35 3.7 20.0 433 20.8 0.001 0.01

Process Condensate Stripper 8.3 0.5 20.0 343 0 0.001 0.01

Flare Stack 35 1.4* 20.0 1273 0.028 0.001 0.01

Gas Turbine Stack 20 3 8.0 753 0.83 0.001 0.01

Auxiliary Boiler Stack 30 3.7 6.0 463 6.39 0.001 0.01

Final 36 Air Quality Impact Assessment

5. Modelling Methodology

5.1 Overview

The modelling used the CSIRO-developed ‘TAPM’ meteorological and air dispersion model (Hurley, 2008a; Hurley et al., 2008). The model was chosen for consistency with previous air quality modelling studies for the Burrup Peninsula completed by CSIRO atmospheric scientists; e.g. Hurley et al. (2004); Physick et al. (2004). The latest version of TAPM (V.4.0.5) was used for the modelling.

The modelling methodology was discussed with EPA Services air quality specialists prior to the commencement of modelling (Jacobs, 2019b). At the EPA Services meeting, it was proposed to use TAPM for the project primarily due to the legacy of TAPM modelling for the Pilbara environment and the need to conduct multiple simulations of annual air emissions scenarios. Subsequent meetings to discuss methodology model development findings, and preliminary outcomes were held with EPA Services and DWER between on 28 March and 13 May 2019. Several aspects of modelling were raised including which version of the TAPM model to be used for the project, and alternative modelling options were discussed. The key outcome of the EPA meetings was that the modelling be demonstrated to be fit-for-purpose. From comparisons of the final TAPM- GRS results with airborne and deposition monitoring results, model accuracy was found to be fit-for-purpose for assessment of the Proposal.

5.2 Burrup Peninsula Modelling Applications

Between 2000 and 2010 the air pollution sources on the Burrup Peninsula and the dispersion of pollutants was a focus of intense study including meteorological modelling, air emissions inventory, and air dispersion modelling. These studies included several TAPM modelling studies by the CSIRO Division of Atmospheric Research, SKM, and other specialist air quality consultants. This section sets out the main findings from a review of those previous studies, important for establishing the modelling methods for this project.

Physick (2001) published a TAPM-Generic Reaction Set (GRS) modelling study on the meteorology and air quality of the Pilbara region, including comparisons with observations at six monitoring sites; this study found: · There was strong seasonal variation of the monthly averaged winds at each site. · There was little difference in the winds between the sites for any given month, especially for wind direction. · Three dominant wind patterns were identified in the coastal region between Karratha and Port Hedland: · An easterly pattern in which winds varied between northeast and southeast over the diurnal period; · A westerly pattern in which the winds varied from northwest to southwest; and · A wind direction rotation anti-clockwise through 360 degrees over 24 hours. · The rotation pattern was assessed as being likely to be important for the recirculation of pollutants, (therefore causing higher air pollutant concentrations around Burrup Peninsula). · The rotation prevailed on some days throughout the year, but more frequently in March, April, August and September.

Apart from the importance of recirculation, Physick (2001) found that emissions from the Burrup Peninsula can meander up the coast to Port Hedland, moving onshore and offshore with sea breezes and nocturnal flows off the land. Thus, in this early phase of studying the atmospheric environment of the Burrup Peninsula, TAPM- GRS was found to be a suitable model to apply to the Pilbara region.

In relation to emissions from the Woodside gas processing facilities, Hurley et al. (2004) determined that buoyancy enhancement of the plumes from the Woodside facilities were important – the effect of plumes combining is to enhance the buoyancy of each individual plume (‘plume merging’). The reactivity of the hydrocarbons known as VOCs emitted from several Woodside facility stacks was found to be important, and

Final 37 Air Quality Impact Assessment

reactivity coefficients for the VOCs were updated. Biogenic emissions were an important consideration, with databases created to address this using a Department of Environmental Protection (DEP) gridded emission inventory (DEP, 2002).

Hurley et al. (2004) advised against assimilation of local wind observations due to the complexity of the region, the sparsity of the wind observations data (two stations only), and local influences such as trees on the wind measurements at Dampier.

Hurley et al. (2008b) reported the following improvements to TAPM V4 over V3: · better performance for a number of annual meteorological verification datasets; · better prediction of wind speed average; · better prediction of temperature standard deviation; · lower root mean square error (RMSE) for all variables; · high index of agreement (IOA) for all variables; and · good prediction of extreme pollution concentrations for several high-quality datasets in regions of varying complexity.

It is noted TAPM version 2.5 was used by the Hurley et al. (2004) study of the Pilbara.

Hurley et al. (2009) provided a summary of some of the improvements in V.4 from V.3: · Land surface parameterisation, nocturnal, low wind conditions, turbulence in the convective boundary layer, “in particular has resulted in improvements in prediction of near surface meteorology.” · Wind and temperature performance for a number of regions of varying complexity—e.g. Kwinana, Kalgoorlie, Perth—"have shown consistently good performance for annual statistics with little mean bias, low RMSE and high IOA.”

In summary, in the 2000s the comparisons of TAPM results with monitoring data indicated TAPM was performing well given the complexity of the coastal meteorology of the Burrup Peninsula region (e.g. Physick et al., 2002), and the complexity of the emissions inventories used (e.g. Hurley et al., 2004).

The previous TAPM modelling and input data used were used as the basis for the modelling for the PU Project detailed in the next section.

5.3 TAPM Model configuration

5.3.1 Main Input Settings

Horizontal and vertical spatial resolution (and time resolution), are key factors that impact on computer speed for a meteorological and air dispersion modelling run. The TAPM modelling for the PU Project drew on previous TAPM set-ups described in this section. Using TAPM, Physick and Blockley (2001) carried out simulations for the Burrup Peninsula with three grids centred near Dampier (each 21 x 21 x 20 grid points), with grid spacings of 10 km, 3 km and 1 km for the meteorology. The grid spacings for the corresponding air quality simulations over the same domains were 5 km, 1.5 km and 0.5 km.

Physick et al. (2004) completed simulations for only for months in the summer (January 1999), winter (July 1998) and the transition season (April 1998). These simulations were carried out on three nests (each 40 x 40 x 20 grid-points) with grid spacings of 30 km, 10 km and 3 km, centred on Karratha. Vertical grid levels were at heights above the ground of 10, 50, 100, 150, 200, 300, 400, 500, 750, 1000, 1250, 1500, 2000, 2500, 3000, 4000, 5000, 6000, 7000 and 8000 m. Terrain elevation was obtained from Geoscience Australia’s gridded 9- second DEM data (approximately 250 m).

Final 38 Air Quality Impact Assessment

For the PU Project, sensitivity tests were undertaken by comparisons of TAPM-predicted winds at Karratha Aerodrome with the Bureau of Meteorology (BoM) measurements of wind speed and wind direction at Karratha Aerodrome and Roebourne. Inclusion of an additional grid with finer horizontal resolution of 400m led to only a small improvement in the accuracy of TAPM-predicted winds. However, the added computational time expense of the additional grid was significant; i.e. weeks, given several scenarios required testing, with many model runs required. As such 1 km resolution modelling was selected for the assessment (meteorological modelling run- times were approximately less than 40 hours for a simulated year). The conclusions of this assessment would not change as a result of the absence of the additional, 400 m resolution, model grid.

Assimilation of local wind observational data was not used in TAPM to enable proper comparisons of results from modelling and monitoring, and to avoid the formation of unrealistic wind vector fields. Hurley et al. (2004) advised that meteorological data assimilation was not advisable for the Burrup Peninsula due to the complexity of the region, the sparsity of (quality) wind data (primarily BoM Karratha Aerodrome), and the local influences on observed wind speeds at Dampier such as trees.

For the current PU Project assessment, a balance between computing speed and accuracy of results was achieved using the TAPM settings set out in Table 5-1.

Table 5-1: Model Configuration

TAPM Modelling Parameter Input data Notes / references

MGA94 co-ordinates: East 470,489 m; North Grid centre coordinates Lat. S. 20° 40’; Long. 116° 43’ 7,714,717 m

Number of grids 3 Grid Spacings (10 km, 3 km, 1 km)

Outer grid spacing 10 km x 10k m Nil

51 (west-east) x 51 (north-south) x 25 Total 2601 ground level grid receptors (inner Number of grid points (vertical) grid).

All defaults as ‘Recommended’ (Hurley, Advanced/Experimental Options Default settings 2008a).

2014 was selected to support model 2014 selected due typical wind pattern as verification of current routine operations determined from analysis of Bureau of Met. against ambient air monitoring records Modelling year Karratha Aerodrome observational data 2010- representative of recent plant ‘full rate’

2018, and good examples of NO2 and O3 operations. 2012 was a back-up year due

measurements at Karratha. good examples of NO2 and O3 measurements, and typical wind pattern.

25 vertical layers including: 10, 50, 100, 150, Vertical Layers (m) 200, 300, 400, 500, 750, 1000, 1250, 1500, 2000… up to 8000 m.

Modelling Mode Air Emissions Scenario Notes:

Tracer mode (mass dispersion) (1) PNO See Section 4.5.2 for scenario descriptions

Tracer mode (mass dispersion) (2) PUC

Tracer mode (mass dispersion) (3) Total NH3

GRS (photochemistry) (4) Baseline

GRS (photochemistry) (5) BPNO

GRS (photochemistry) (6) BPUC

GRS (photochemistry) (7) FPNO

Particulate mode (PM2.5 and PM10) (8) PNO particulate urea PM2.5/PM10 ratio 30% (GHD, 2009)

Final 39 Air Quality Impact Assessment

5.3.2 Land Use

TAPM uses terrain elevations and land use data to describe the geography of a study area that underlies the fields of three-dimensional meteorological data computed and allowed to evolve over the modelled study area. Land use data include parameters important for boundary layer meteorological computations, where the meteorology makes contact with the land surface. One of these parameters is surface roughness, which influences turbulence in the atmospheric boundary layer or mixing layer, which in turn influences the dispersion of air pollutants.

Parameters for vegetation types defined in the TAPM model are set out in Table 5-2 (Hurley 2008a).

Table 5-2: TAPM Vegetation Characteristics

Surface fraction Minimum stomatal Type Height (m) Leaf Area Index -1 (sf) resistance (s )

Forest - low dense 9.00 0.75 3.9 200

Shrubland - tall mid-dense scrub 3.00 0.50 2.6 160

Shrubland - low mid-dense 1.00 0.50 1.4 90

Shrubland - low sparse 0.60 0.25 1.5 90

Grassland - mid-dense tussock 0.60 0.50 1.2 80

Pasture mid-dense 0.45 0.50 1.2 40

Urban and Industrial 10.00 0.75 2.0 100

The TAPM land use settings for the Burrup Peninsula were based on those of Physick and Blockley (2001). For the 1 km grid, land-use classification in the data set accompanying the TAPM modelling package was changed from a land category to water for grid points corresponding to the Dampier Salt Farm at the lower end of the Burrup Peninsula. A roughness length of 0.9 m was assigned to Burrup Peninsula grid points by changing the land-use category in that region to low dense forest, which simulates the rough rocky landscape. The final two nested grids (3 km and 1 km) used for the modelling are illustrated in the image extracts from the TAPM Graphical User Interface in Figure 5-1.

Final 40 Air Quality Impact Assessment

TAPM 3km grid – terrain and water TAPM 1km grid - terrain and water

TAPM 3km grid (modified) – vegetation & land use TAPM 1km grid (modified) – vegetation & land use

Figure 5-1: TAPM 3km and 1km Grids – Terrain, Vegetation and Land Use

Final 41 Air Quality Impact Assessment

5.3.3 Deep Soil Moisture Content

Estimates for monthly varying Deep Soil Moisture Content (DSMC) were interpolated linearly based on tests by Physick et al. (2004) that showed best agreement with wind data obtained using: DSMC 0.05 m3 m-3 for January and April; and DSMC 0.15 m3 m-3 for July. The modified DSMC values used for the modelling assessment are shown in Figure 5-2 .

Figure 5-2: Deep Soil Moisture Content Settings

5.3.4 Photochemical Modelling

TAPM’s in-built photochemical modelling scheme was used for this modelling assessment for consistency with previous CSIRO and SKM modelling studies. In TAPM, gas-phase photochemical modelling is based on the Generic Reaction Set (GRS) semi-empirical mechanism of Azzi et al. (1992) and the hydrogen peroxide modification of Venkatram et al. (1997). TAPM also includes gas-phase and aqueous-phase reactions of SO2 and particles. Aqueous-phase reactions were based on Seinfeld and Pandis (2016).

TAPM simulates 10 chemical reactions for 13 species in GRS mode including: smog reactivity (Rsmog), the radical pool (RP), hydrogen peroxide (H2O2), nitric oxide (NO), NO2, O3, SO2. Further details are provided in Hurley (2008a).

More complex photochemical modelling could be undertaken for the Burrup Peninsula; e.g., using TAPM-CTM (Cope and Lee, 2009). However, the selection of TAPM-GRS provided an appropriate balance between model accuracy (as determined by comparisons with monitoring results) and computational time cost. The use of TAPM-GRS also allowed for the efficient modelling of multiple year-long simulations, a feature important to make comparisons between annual averages for each scenario.

Comparisons of TAPM-GRS results with monitoring data obtained on the Burrup Peninsula, were the key tests of model accuracy. The current application of TAPM-GRS to the Pilbara indicated the most substantial gains towards model accuracy were through improvements to the air emissions inventories used as input.

Using the previous CSIRO studies as the main foundational guides, inputs required at the user interface for the photochemical modelling included the following estimates for background air pollutant levels: NOx (1 ppb),

Final 42 Air Quality Impact Assessment

background smog reactivity or the so-called ‘Rsmog’ parameter (0.2 ppb), and background O3 (25 ppb). Values for Rsmog were calculated for every modelled source using estimates for the total VOC emission rate (g/s) and an estimate of reactivity associated with the source type. Air Assessments (2010b) stated that generally it is the boundary (background) condition of Rsmog that is most important, with 'surface sources contributing little Rsmog’. Initially the estimate for background Rsmog (0.2 ppb) was selected by Hurley et al. (2004).

TAPM also allows for the input of large-scale area emissions of air pollutants to include as background. Again, using the previous CSIRO studies as a guide, the CSIRO biogenic emissions databases used with TAPM are illustrated in Figure 5-3 (NOx), and Figure 5-4 (Rsmog). The figures are overlaid on the base map image of the Burrup Peninsula study area, representing the TAPM inner-grid.

Figure 5-3: CSIRO Biogenic NOx Area Emissions Database and Current Study Area (Inset)

Final 43 Air Quality Impact Assessment

Figure 5-4: CSIRO Biogenic Area Rsmog Emissions Database and Current Study Area (Inset)

Another area source file used with previous TAPM modelling included emissions from shipping and the relatively small townships of Dampier (population approximately 1,100), and Karratha (population approximately 15,800). A weakness of this database was overestimating the effects of the shipping emissions by excluding the effects of hot (buoyant) exhausts from ship engines. This weakness in the emissions estimates was recognised by Air Assessments (2010). For this project, the effects of shipping were modelled by including ship engines running continuously throughout a year at every available berth in the Burrup Peninsula and Cape Lambert.

Area emissions from Dampier and Karratha were also excluded from the modelling because the small amounts of emissions from road traffic from these towns were insignificant relative to the industrial sources. In any case, by including background levels of NOx, O3, particles and hydrocarbons in the modelling, the emissions from Dampier and Karratha were included implicitly.

5.3.5 Selection of Year for Modelling

The TAPM meteorological simulation year 2014 was selected as the basis for the air quality assessment supporting the PU Project. The process for selecting this representative year included a review of 9 years of hourly-average meteorological observations data from BoM Karratha Aerodrome (2010-2018). Annual statistics for wind speed and wind direction were examined for any annual meteorological variations in the Burrup region.

Final 44 Air Quality Impact Assessment

This included a review of cyclones in the Pilbara to check the potential effects on Karratha wind speed (Appendix B).

The completeness and representativeness of air quality monitoring data was considered. The selection for the simulation was 2014, which was considered to be representative of meteorological conditions, combined with an annual air quality monitoring dataset that best represented the existing industrial air emissions situation.

PLP was commissioned in 2012, ramped up in the later half 2012, and was at full production in 2013, although with some variability in the 2013 operations. The year 2014 was determined to be a good record of high KGP and PLP production rates and overlapped with a solid ambient air quality monitoring record. All factors combined, the year 2014 was selected as the best meteorological simulation year for TAPM.

TAPM was used to produce modelling results for wind speed and wind direction for 2014. The predicted meteorological outputs were compared with the 2014 hourly datasets from the Bureau of Meteorology (BoM) weather stations at Karratha, Roebourne and Legendre Island to assess the model’s suitability for dispersion modelling. This comparison is outlined in Appendix C.

A further analysis of meteorological modelling performance is described in Section 5.6.

5.4 Modelled Variable Background Particulate Matter

As mentioned in Section 3.2, levels of PM10 and PM2.5 in the Pilbara are affected by smoke from bushfires, dust storms, sea spray, for example. This is a similar situation to many other parts of Australia. Also, in the Pilbara, there are dust emissions from industrial sources due to materials handling such as occurs at the ship-loading berths around the Burrup Peninsula and Cape Lambert.

A detailed assessment of dust emissions associated with the ports and other sources was outside the scope of this assessment – the PU Project comprises only two PM10 sources of 5.4 g/s each (maximum; see Table 4-12). This represents a relatively small contribution to the Pilbara air environment. However, PM10 and PM2.5 emissions were included in the modelling for the PU Project, with emission estimates determined where possible for all the sources modelled (see Section 4).

Also, the hourly average background PM10 and PM2.5 was modelled for the PU Project and input to TAPM-GRS. The long-term record of PM10 measured by Hamersley Iron over 2002-2006 (see Figure 3-1), motivated the creation of a database of seasonally cyclic, hourly-varying PM10 and PM2.5 to use for the PU Project. A sinusoidal fit was determined for the PM10 measurements from 2002-2006, constrained by an estimated 3 3 minimum PM10 concentration during the dry season (mid-year) of 15 µg/m , and a maximum of 35 µg/m for the wet season (time zero was set to the first hour of each year). The Hamersley Iron data (Figure 3-1) were unavailable for analysis, so the resulting PM10 model is shown with more recent PM10 results calculated from an analysis of results from provided by Yara (2017); see Figure 5-5. The Yara (2017) data includes some peaks, as the monitoring locations TRA-1 and TRA-2 were located around the Yara TAN plant site boundaries; these monitoring stations are shown in Appendix A.

Final 45 Air Quality Impact Assessment

Figure 5-5: Modelled Seasonally-Cyclic Background PM10 and Yara (2017) Measurements

Inspection of the results shown in Figure 5-5 indicates the modelled background PM10 is producing reasonable results for the Yara site, even though the Yara data were obtained from near the Yara industrial facilities. It is emphasised the sinusoid attempts to model the background or baseline PM10; i.e., not emissions from the industrial sources. The lower PM10 levels during mid-year were found to be due to lower wind speeds occurring in the dry season, and higher levels of PM10 observed during the build-up to the wet season were due to increased wind speeds. Apart from industrial emissions, peaks in the PM10 data shown in the Hamersley Iron data (Figure 3-1), and other similar results for the Pilbara, would have included significant contributions due to emissions from fires, dust storms, and, being a coastal site, salt particles from sea spray would have contributed to the measured PM10 levels.

The PM2.5 was calculated also by the cyclic-seasonal model, using a single estimate of the PM2.5/PM10 ratio. The PM2.5/PM10 ratio was assumed to be 25% for all hours, which was considered to be conservative (high) for the higher PM10 concentrations, because higher PM10 concentrations are usually associated with raised dust with lower PM2.5/PM10 ratios; see Section 3.2.

5.5 Deposition of Gaseous Pollutants

The deposition flux of nitrogen and sulfur on Burrup Peninsula may be relevant for effects on rock art (DWER. 2019); a summary of results for the gaseous deposition components obtained from measurements was set out in Section 3.7. In the absence of accepted standards for potential impacts on rock art (DWER, 2019), TAPM- GRS outputs were obtained for the NO2 and SO2 dry deposition for the purpose of comparisons with measurements only. It is emphasised an assessment of the impacts on rock art is outside the scope of this assessment.

TAPM does not provide outputs for the deposition of NH3, so (dry) deposition of NH3 was calculated from the model results for annual average airborne concentrations of NH3 combined with an estimate of the fall velocity for the molecule (0.60 cm/s); e.g., see Shen et al. (2016). Some calculated results also included an estimate for background airborne NH3 (see Section 3.3), which were constrained in part by measurements that showed the 2 background NH3 dry deposition was approximately 4 meq/m /year (see Section 3.7).

Final 46 Air Quality Impact Assessment

The model results for deposition were illustrated as contour plots in a similar way to the standard presentation of results for (airborne) GLCs. Results are provided to enable comparisons with monitoring results such as those from Gillett (2008), Gillett et al. (2012), a CSIRO summary of results obtained by Gillett (2014); and the Woodside (2019) summary of results.

It is noted the TAPM calculations for dry and wet deposition of NO2 and SO2, which are detailed in Hurley (2008), use a similar method to that adopted by Gillett (2008). The results may differ slightly between the methods depending on parameters such as, deposition velocities of the gases, and various resistance parameters used in the calculations by each study. Measured airborne concentrations are used to calculate dry deposition of a gas. Variability in the input parameters of approximately 10% (Gillett, 2008), means the TAPM calculations of deposition could differ from the ‘measured’ values by approximately 10% or slightly greater.

The conversion of the TAPM results for deposition (D) of gaseous molecules in units of mg/m2/year to meq/m2/year was calculated using the equation, D = m/M × z, where m is the deposition mass (mg) predicted by TAPM, M is the molecular mass; e.g., 46 g/mol for NO2, and z is the charge on the ion produced. Note the deposition flux calculated in this way is equal to the number of moles of the acid ion (H+) assumed to be activated in liquid water by the ion; e.g., for NO2, z is assigned a value of one given all the deposited NO2 is - assumed to convert to the nitrate ion (NO3 ) with its single negative charge; i.e., nitric acid.

A summary of the data used for the dry deposition calculations is provided in Table 5-3.

Table 5-3: Gaseous Dry Deposition Data Summary

Parameter NH3 NO2 SO2

Mass example, milligram (mg) 1 1 1

Molar mass (g/mole) 17.03 46.0 64.06

+ - - 2- Charge in terms of potential to create H 1 e.g. as NH2 1 e.g. as NO3 2 e.g. as SO4

Equivalent example, milliequivalents (meq) 0.0587 0.0217 0.0312

5.6 Meteorological Modelling Performance

As is already known and reported, the TAPM simulations underestimated wind speed for the 2014 meteorological simulation case (Section 5.3.5; Appendix C). In addition, performance tests were undertaken for meteorological simulations for the Burrup Peninsula, for 2012 and 2018 (Appendix D). The correlation between the modelled and measured wind speeds for the 2012 simulation was particularly poor, due to particularly bad correlation from February to May in 2012. However, the basic statistics of the 2012 simulation were similar to those for the 2014 simulation. Other tests undertaken indicate the current TAPM predictions for wind speed on Burrup Peninsula did not perform as well as the 1999 meteorological simulation generated by the CSIRO; e.g., Hurley et al. (2004).

The current TAPM results for temperature are very good (2012, 2014 and 2018), and on a par with the CSIRO simulations of 1999 meteorology for Burrup Peninsula. The current results for relative humidity are satisfactory, although TAPM did not perform as well for the 2014 simulation. The TAPM 2014 and 2018 meteorological simulations for Burrup Peninsula are considered satisfactory for the purpose of air quality impact assessment. The poor correlation of wind speeds in the 2012 simulation means further review of that simulation would be advisable, prior to its use in an assessment (see Appendix D for more details).

There are sound, scientific reasons for using the latest version of the model. TAPM V.4 included a substantial number of new improvements to the modelled physics; e.g., see Hurley et al. (2009), and Section 5.2 of this report.

The full details of the performance tests undertaken, and the results, are provided in Appendix D.

Final 47 Air Quality Impact Assessment

6. Results

6.1 Overview

This section provides the model results for predicted concentrations and deposition of pollutants. Contour plots for each species and averaging period enable comparisons between the modelled results and air quality standards and provide an indication of the dispersion pattern for each air pollutant over the course of a year.

The results in this section are provided in order of substance and assessment parameter for each scenario. Comparisons are provided with standards where relevant. A summary of the results is provided Table 6-1 (contour plots are not provided for cases where the dispersion patterns are identical or similar to those for related scenarios).

Table 6-1: TAPM-GRS Results Presented as Contour Plots

Scenario Results for PM Results for Gases Results for Deposition

(1) Perdaman in isolation; Maximum 24-hour PM10 Max. 1h NH3 Annual NH3 Normal Operations (PNO) Annual average PM10 Max. 1h NO2 Annual NO2 (kg/ha/year)

Maximum 24-hour PM2.5 Max. 1h SO2 Annual SO2 (kg/ha/year)

Annual average PM2.5 Max. 1h CH2O Max. 1h Methanol

(2) Perdaman in isolation; Maximum 24-hour PM10 Max. 1h NH3 N/A Upset Conditions (PUC) Maximum 24-hour PM2.5 Max. 1h NO2

Max. 1h SO2

Max. 1h CH2O Max. 1h Methanol

(3) Perdaman normal N/A Max. 1h NH3 N/A operations plus Yara Pilbara

sources (NH3 only) (Total Ammonia)

(4) Existing air emissions Maximum 24-hour PM10 Max. 1h NH3 Annual NH3 (scenario 1 plus

scenario (Baseline) estimate for background NH3) Annual average PM10 Max. 1h & annual NO2

Annual NO2 (kg/ha/year) Maximum 24-hour PM2.5 Max. 1h & 4h O3

Annual SO2 (kg/ha/year) Annual average PM2.5 Max. 1h, max. 24h & annual SO2

(5) Baseline with PNO Maximum 24-hour PM10 Max. 1h NH3 Annual NH3 (scenario 1 plus

(BPNO) estimate for background NH3) Annual average PM10 Max. 1h & annual NO2

Annual NO2 (kg/ha/year) Maximum 24-hour PM2.5 Max. 1h & 4h O3

Annual SO2 (kg/ha/year) Annual average PM2.5 SO2 results – similar to (3)

(6) Baseline with PUC Maximum 24-hour PM10 Max. 1h NH3 N/A (BPUC) Maximum 24-hour PM2.5 Max. 1h NO2

Max. 1h & 4h O3

(7) Baseline with PNO and Maximum 24-hour PM10 Max. 1h NH3 Annual NH3 (scenario 1 plus

other proposed future projects estimate for background NH3) Annual average PM10 Max. 1h & annual NO2 (FPNO) Annual NO2 (kg/ha/year) Maximum 24-hour PM2.5 Max. 1h & 4h O3

Annual SO2 (kg/ha/year) Annual average PM2.5 SO2 results – similar to (3)

Final 48 Air Quality Impact Assessment

6.2 Particulate Matter as PM10

This section provides the TAPM results for airborne PM10 concentrations for comparisons with air quality standards for the protection of human health.

6.2.1 Scenario PNO – Maximum 24h PM10 GLC

Figure 6-1: PNO – Maximum 24h PM10 GLC (µg/m3) · Perdaman in isolation – normal operations (not a cumulative result i.e. not for assessment). · Maximum grid receptor concentration, 12.4 µg/m3. · NEPM (Ambient Air Quality) standard, 50 µg/m3 (for information only).

Final 49 Air Quality Impact Assessment

6.2.2 Scenario PUC – Maximum 24h PM10 GLC

Figure 6-2: PUC – Maximum 24h PM10 GLC (µg/m3) · Perdaman in isolation – upset conditions (not a cumulative result i.e. not for assessment). · Maximum grid receptor concentration, 19.1 µg/m3. · NEPM (Ambient Air Quality) standard, 50 µg/m3 (for information only).

Final 50 Air Quality Impact Assessment

6.2.3 Scenario Baseline – Maximum 24h PM10 GLC

Figure 6-3: Baseline – Maximum 24h PM10 GLC (µg/m3) 3 · Maximum grid receptor concentration, 35.5 µg/m (including modelled background varying PM10). · NEPM (Ambient Air Quality) standard, 50 µg/m3. · Result of cumulative air quality impact assessment: no exceedances.

Final 51 Air Quality Impact Assessment

6.2.4 Scenario BPNO – Maximum 24h PM10 GLC

Figure 6-4: BPNO – Maximum 24h PM10 GLC (µg/m3) 3 · Maximum grid receptor concentration, 44.7 µg/m (including modelled background varying PM10). · NEPM (Ambient Air Quality) standard, 50.0 µg/m3. · Result of cumulative air quality impact assessment: no exceedances.

Final 52 Air Quality Impact Assessment

6.2.5 Scenario BPUC – Maximum 24h PM10 GLC

Figure 6-5: BPUC – Maximum 24h PM10 GLC (µg/m3) 3 · Maximum grid receptor concentration, 53.0 µg/m (including modelled varying-background PM10). · NEPM (Ambient Air Quality) standard, 50.0 µg/m3. · Result of cumulative air quality impact assessment: one exceedance (one grid point only in close proximity to PU Project, of 2601 grid points).

Final 53 Air Quality Impact Assessment

6.2.6 Scenario FPNO – Maximum 24h PM10 GLC

Figure 6-6: FPNO – Maximum 24h PM10 GLC (µg/m3) 3 · Maximum grid receptor concentration, 44.6 µg/m (including modelled varying-background PM10). · NEPM (Ambient Air Quality) standard, 50 µg/m3. · Result of cumulative air quality impact assessment: no exceedances.

Final 54 Air Quality Impact Assessment

6.2.7 Scenario PNO – Annual Average PM10 GLC

Figure 6-7: PNO – Annual Average PM10 GLC (µg/m3) · Perdaman in isolation – normal operations (not a cumulative result i.e. not for assessment). · Maximum grid receptor concentration, 6.4 µg/m3; includes capacity factor 90.6%. · NEPM (Ambient Air Quality) standard, 25 µg/m3 (for information only).

Note: there are no annual average PM10 results for the PUC scenario, which is a short-term emissions scenario.

Final 55 Air Quality Impact Assessment

6.2.8 Scenario Baseline – Annual Average PM10 GLC

Figure 6-8: Baseline – Annual Average PM10 GLC (µg/m3) · Maximum grid receptor concentration, 24.8 µg/m3. · NEPM (Ambient Air Quality) standard, 25 µg/m3. · Result of cumulative air quality impact assessment: no exceedances.

Final 56 Air Quality Impact Assessment

6.2.9 Scenario BPNO – Annual Average PM10 GLC

Figure 6-9: BPNO – Annual Average PM10 GLC (µg/m3) · Maximum grid receptor concentration, 30.9 µg/m3. · NEPM (Ambient Air Quality) standard, 25 µg/m3. · Some exceedances in small part of study area due high background; worst case PU Project contribution 5.6 µg/m3.

Final 57 Air Quality Impact Assessment

6.2.10 Scenario FPNO – Annual Average PM10 GLC

Figure 6-10: FPNO – Annual Average PM10 GLC (µg/m3) · Maximum grid receptor concentration, 30.8 µg/m3. · NEPM (Ambient Air Quality) standard, 25 µg/m3. · Some exceedances in close proximity of PU Project but mainly due to high background; worst case PU Project contribution 6.4 µg/m3 including capacity factor of 90.6%.

Final 58 Air Quality Impact Assessment

6.3 Particulate Matter as PM2.5

This section provides the TAPM results for airborne PM2.5 concentrations for comparisons with air quality standards for the protection of human health.

6.3.1 Scenario Baseline – Maximum 24h PM2.5 GLC

Figure 6-11: Baseline – Maximum 24h PM2.5 GLC (µg/m3) · Maximum grid receptor concentration, 15.5 µg/m3. · NEPM (Ambient Air Quality) standard, 25 µg/m3. · Result of cumulative air quality impact assessment: no exceedances.

Final 59 Air Quality Impact Assessment

6.3.2 Scenario BPNO – Maximum 24h PM2.5 GLC

Figure 6-12: BPNO – Maximum 24h PM2.5 GLC (µg/m3) · Maximum grid receptor concentration, 17.4 µg/m3. · Using the PNO results, the PU Project contribution to this result is estimated to be 3.2 µg/m3. · NEPM (Ambient Air Quality) standard, 25 µg/m3. · Result of cumulative air quality impact assessment: no exceedances.

Final 60 Air Quality Impact Assessment

6.3.3 Scenario BPUC – Maximum 24h PM2.5 GLC

Figure 6-13: BPUC – Maximum 24h PM2.5 GLC (µg/m3) · Maximum grid receptor concentration, 18.9 µg/m3. · Using the PUC results, the PU Project contribution to this result is estimated to be 5.7 µg/m3. · NEPM (Ambient Air Quality) standard, 25 µg/m3. · Result of cumulative air quality impact assessment: no exceedances.

Final 61 Air Quality Impact Assessment

6.3.4 Scenario FPNO – Maximum 24h PM2.5 GLC

Figure 6-14: FPNO – Maximum 24h PM2.5 GLC (µg/m3) · Maximum grid receptor concentration, 17.4 µg/m3. · Using the PNO results, the PU Project contribution to this result is estimated to be 3.7 µg/m3. · NEPM (Ambient Air Quality) standard, 25 µg/m3. · Result of cumulative air quality impact assessment: no exceedances.

Final 62 Air Quality Impact Assessment

6.3.5 Scenario Baseline – Annual Average PM2.5 GLC

Figure 6-15: Baseline – Annual Average PM2.5 GLC (µg/m3) · Maximum grid receptor concentration, 8.4 µg/m3. · NEPM (Ambient Air Quality) standard, 8 µg/m3.

· Some exceedances primarily due to high background PM2.5, concentrated around the modelled emissions from the shipping berths. Note these results do not include PU Project emissions.

Final 63 Air Quality Impact Assessment

6.3.6 Scenario BPNO – Annual Average PM2.5 GLC

Figure 6-16: BPNO – Annual Average PM2.5 GLC (µg/m3) · Maximum grid receptor concentration, 10.3 µg/m3. · NEPM (Ambient Air Quality) standard, 8 µg/m3. · Some exceedances of the standard centred around the industrial area. · Using the worst case PNO results the PU Project contribution to the maximum is 1.4 µg/m3.

Final 64 Air Quality Impact Assessment

6.3.7 Scenario FPNO – Annual Average PM2.5 GLC

Figure 6-17: FPNO – Annual Average PM2.5 GLC (µg/m3) · Maximum grid receptor concentration, 10.3 µg/m3. · NEPM (Ambient Air Quality) standard, 8 µg/m3. · Some exceedances of the standard centred around the industrial area. · The PU Project contribution to the maximum is estimated to be 1.9 µg/m3.

Final 65 Air Quality Impact Assessment

6.4 Ammonia

This section provides the TAPM (tracer mode) results for airborne NH3 concentrations for comparisons with air quality standards for the protection of human health.

6.4.1 Scenario PNO – Maximum 1h NH3 GLC

Figure 6-18: PNO – Maximum 1h NH3 GLC (µg/m3) · Perdaman in isolation – normal operations (PNO). · Maximum grid receptor concentration, 76.9 µg/m3. · NSW EPA impact assessment criterion, 330 µg/m3. · Result of assessment – no exceedences.

Final 66 Air Quality Impact Assessment

6.4.2 Scenario PUC – Maximum 1h NH3 GLC

Figure 6-19: PUC – Maximum 1h NH3 GLC µg/m3 · Perdaman in isolation – upset conditions (PUC). · Maximum grid receptor concentration, 75.9 µg/m3 (less than previous result for PNO). · NSW EPA impact assessment criterion, 330 µg/m3 (for information only; not for assessment). · Result of assessment – no exceedences.

Final 67 Air Quality Impact Assessment

6.4.3 Scenario Total Ammonia – Maximum 1h NH3 GLC

The results illustrated in this sub-section for the scenario ‘Total Ammonia’ represent worst-case results covering the scenarios BPNO, BPUC, and FPNO – the results include NH3 emissions estimates for the PU Project and the Yara plants, plus an estimate for constant background NH3.

Figure 6-20: Total Ammonia – Maximum 1h NH3 GLC µg/m3

· Total Ammonia: Perdaman, Yara Pilbara ammonium nitrate and fertiliser plants, and NH3 background. · Maximum grid receptor concentration, 77.3 µg/m3 (only slightly higher than the maximum for PNO).

3 · Includes estimate for constant background NH3, 0.35 µg/m (see Section 3.3). · NSW EPA impact assessment criterion, 330 µg/m3. · Result of assessment – no exceedances.

Final 68 Air Quality Impact Assessment

6.5 Nitrogen Dioxide

This section provides the TAPM results for airborne NO2 concentrations for comparisons with air quality standards for the protection of human health.

6.5.1 Scenario Baseline – Maximum 1h NO2 GLC

Figure 6-21: Baseline – Maximum 1h NO2 GLC (ppb) · Maximum grid receptor concentration, 42.6 ppb. · NEPM (Ambient Air Quality) standard, 120 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 69 Air Quality Impact Assessment

6.5.2 Scenario BPNO – Maximum 1h NO2 GLC

Figure 6-22: BPNO – Maximum 1h NO2 GLC (ppb) · Maximum grid receptor concentration, 43.1 ppb. · NEPM (Ambient Air Quality) standard, 120 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 70 Air Quality Impact Assessment

6.5.3 Scenario BPUC – Maximum 1h NO2 GLC

Figure 6-23: BPUC – Maximum 1h NO2 GLC (ppb) · Maximum grid receptor concentration, 42.9 ppb. · NEPM (Ambient Air Quality) standard, 120 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 71 Air Quality Impact Assessment

6.5.4 Scenario FPNO – Maximum 1h NO2 GLC

Figure 6-24: FPNO – Maximum 1h NO2 GLC (ppb) · Maximum grid receptor concentration, 43.9 ppb. · NEPM (Ambient Air Quality) standard, 120 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 72 Air Quality Impact Assessment

6.5.5 Scenario Baseline – Annual Average NO2 GLC

Figure 6-25: Baseline – Annual Average NO2 GLC (ppb) · Maximum grid receptor concentration, 5.0 ppb. · NEPM (Ambient Air Quality) standard, 30 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 73 Air Quality Impact Assessment

6.5.6 Scenario BPNO – Annual Average NO2 GLC

Figure 6-26: BPNO – Annual Average NO2 GLC (ppb) · Maximum grid receptor concentration, 5.6 ppb. · NEPM (Ambient Air Quality) standard, 30 ppb. · Result of cumulative air quality impact assessment: no exceedances

Final 74 Air Quality Impact Assessment

6.5.7 Scenario FPNO – Annual Average NO2 GLC

Figure 6-27: FPNO – Annual Average NO2 GLC (ppb) · Maximum grid receptor concentration, 5.9 ppb. · NEPM (Ambient Air Quality) standard, 30 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 75 Air Quality Impact Assessment

6.6 Ozone

This section provides the TAPM results for airborne O3 concentrations for comparisons with air quality standards for the protection of human health.

6.6.1 Scenario Baseline – Maximum 1h O3 GLC

Figure 6-28: Baseline – Maximum 1h O3 GLC (ppb) · Maximum grid receptor concentration, 61.8 ppb. · NEPM (Ambient Air Quality) standard, 100 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 76 Air Quality Impact Assessment

6.6.2 Scenario BPNO – Maximum 1h O3 GLC

Figure 6-29: BPNO – Maximum 1h O3 GLC (ppb) · Maximum grid receptor concentration, 62.0 ppb. · NEPM (Ambient Air Quality) standard, 100 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 77 Air Quality Impact Assessment

6.6.3 Scenario BPUC – Maximum 1h O3 GLC

Figure 6-30: BPUC – Maximum 1h O3 GLC (ppb) · Maximum grid receptor concentration, 61.9 ppb. · NEPM (Ambient Air Quality) standard, 100 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 78 Air Quality Impact Assessment

6.6.4 Scenario FPNO – Maximum 1h O3 GLC

Figure 6-31: FPNO – Maximum 1h O3 GLC (ppb) · Maximum grid receptor concentration, 63.0 ppb. · NEPM (Ambient Air Quality) standard, 100 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 79 Air Quality Impact Assessment

6.6.5 Scenario Baseline – Maximum 4h O3 GLC

Figure 6-32: Baseline – Maximum 4h O3 GLC (ppb) · Maximum grid receptor concentration, 58.2 ppb. · NEPM (Ambient Air Quality) standard, 80 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 80 Air Quality Impact Assessment

6.6.6 Scenario BPNO – Maximum 4h O3 GLC

Figure 6-33: BPNO – Maximum 4h O3 GLC (ppb) · Maximum grid receptor concentration, 58.3 ppb. · NEPM (Ambient Air Quality) standard, 80 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 81 Air Quality Impact Assessment

6.6.7 Scenario BPUC – Maximum 4h O3 GLC

Figure 6-34: BPUC – Maximum 4h O3 GLC (ppb) · Maximum grid receptor concentration, 58.2 ppb. · NEPM (Ambient Air Quality) standard, 80 ppb. · Result of cumulative air quality impact assessment: no exceedences.

Final 82 Air Quality Impact Assessment

6.6.8 Scenario FPNO – Maximum 4h O3 GLC

Figure 6-35: FPNO – Maximum 4h O3 GLC (ppb) · Maximum grid receptor concentration, 59.7 ppb. · NEPM (Ambient Air Quality) standard, 80 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 83 Air Quality Impact Assessment

6.7 Other Air Pollutants

This section provides TAPM results for airborne concentrations for SO2, methanol and formaldehyde, for comparisons with air quality standards for the protection of human health.

6.7.1 Sulfur Dioxide

The SO2 emission rates varied by very little between the scenarios. As such only one set of contour plots is provided–for the Baseline case, which is representative of all model scenarios.

Summaries of the model grid point maxima for the SO2 GLCs, and corresponding NEPM standards, is provided in Table 6-2 (maximum 1-hour averages), Table 6-3 ( maximum 24-hour averages) and Table 6-4 (annual averages). These results show that the relevant NEPM standards are not expected to be exceeded anywhere in the study area.

Table 6-2: Maximum 1-hour Average SO2 GLC (ppb)- Sensitive Receptors and Grid Maxima

Receptor PNO PUC BASELINE BPNO BPUC FPNO

Grid max. 0.17 0.13 18.2 18.2 18.2 18.1

NEPM Standard 200 200 200 200 200 200

Table 6-3: Maximum 24-hour Average SO2 GLC (ppb)- Sensitive Receptors and Grid Maxima

Receptor PNO PUC BASELINE BPNO BPUC FPNO

Grid max. 0.05 0.11 7.0 7.0 7.0 7.0

NEPM Standard 80 80 80 80 80 80

Table 6-4: Annual Average SO2 GLC (ppb)- Sensitive Receptors and Grid Maxima

Receptor PNO PUC BASELINE BPNO BPUC FPNO

Grid max. 0.02 N/A 4.5 4.5 N/A 4.5

NEPM Standard 20 20 20 20 20 20

Final 84 Air Quality Impact Assessment

6.7.2 Scenario Baseline – Maximum 1h SO2 GLC

Figure 6-36: Baseline – Maximum 1h SO2 GLC (ppb) · Maximum grid receptor concentration, 18.2 ppb. · NEPM (Ambient Air Quality) standard, 200 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 85 Air Quality Impact Assessment

6.7.3 Scenario Baseline – Maximum 24h SO2 GLC

Figure 6-37: Baseline – Maximum 24h SO2 GLC (ppb) · Maximum grid receptor concentration, 7.0 ppb. · NEPM (Ambient Air Quality) standard, 80 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 86 Air Quality Impact Assessment

6.7.4 Scenario Baseline – Annual Average SO2 GLC

Figure 6-38: Baseline – Annual Average SO2 GLC (ppb) · Maximum grid receptor concentration, 4.5 ppb. · NEPM (Ambient Air Quality) standard, 20 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 87 Air Quality Impact Assessment

6.7.5 Formaldehyde

Formaldehyde was eliminated from detailed modelling and assessment as a low-risk substance; maximum hourly average GLCs due to the PU Project were expected to be approximately 1 µg/m3, well below the assessment criterion of 20 µg/m3; see also Section 3.5.

6.7.6 Methanol

Methanol was eliminated from detailed modelling and assessment as a low-risk substance; maximum hourly average GLCs due to the Perdaman proposal were expected to be approximately 1 µg/m3, very much below the assessment criterion of 3000 µg/m3.

6.8 Potential Effects on Vegetation

6.8.1 Summary

The purpose of this section is to provide the results of an assessment on the potential effects on vegetation health due to airborne NOx and SO2 emissions. The assessment results for the PU Project scenarios are listed in Table 6-5. Results for the new PU Project scenarios for annual NOx are provided in Figure 6-41 (BPNO) and Figure 6-42 (FPNO); for annual SO2 results, see the Baseline results in Figure 6-38.

In summary, for the assessment of effects on vegetation:

· Maximum results for annual average NOx (ppb) for all scenarios are less than the relevant EU (2008) standard of 16 ppb.

· Maximum results for annual average SO2 (ppb) for all scenarios are less than the relevant EU (2008) standard of 8 ppb.

For all scenarios the highest SO2 concentrations were predicted for locations adjacent to the shipping berths, which were conservatively modelled as continuous sources. An Australian Maritime Safety Authority (AMSA) requirement to reduce the sulfur content of fuel for shipping will lower the future risk of impact on vegetation from SO2 emissions, which is due to come in force 1 January 2020 (AMSA, 2018).

Table 6-5: Maximum Results for Annual NOx and SO2 GLC (ppb)

Assessment Parameter & EU PNO Baseline BPNO FPNO (2008) Standard

Annual NOx 2.4 7.7 8.6 9.0 16.2 ppb

Annual SO2 0.05 4.5 4.5 4.5 7.8 ppb

3 o 3 o Notes: Annual NOx standard (16.2 ppb) calculated from 30 µg/m as NO2 at 30 C. Annual SO2 (7.8 ppb) calculated from 20 µg/m at 30 C.

Final 88 Air Quality Impact Assessment

6.8.2 Scenario PNO – Annual Average NOX GLC

Figure 6-39: PNO – Annual Average NOX GLC (ppb) · Perdaman in isolation – normal conditions (not a cumulative result i.e. not for assessment). · Maximum grid receptor concentration, 0.5 µg/m3 (from TAPM in tracer mode). · EU (2008) standard for protection of vegetation, 30 µg/m3 (for information only).

Final 89 Air Quality Impact Assessment

6.8.3 Scenario Baseline – Annual Average NOX GLC

Figure 6-40: Baseline – Annual Average NOX GLC (ppb) · Maximum grid receptor concentration, 7.7 ppb (from TAPM in GRS mode). · EU (2008) standard for protection of vegetation, 16.2 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 90 Air Quality Impact Assessment

6.8.4 Scenario BPNO – Annual Average NOX GLC

Figure 6-41: BPNO – Annual Average NOX GLC (ppb) · Maximum grid receptor concentration, 8.6 ppb (from TAPM in GRS mode). · EU (2008) standard for protection of vegetation, 16.2 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 91 Air Quality Impact Assessment

6.8.5 Scenario FPNO – Annual Average NOx GLC

Figure 6-42: FPNO – Annual Average NOx GLC (ppb) · Maximum grid receptor concentration, 9.1 ppb (TAPM in GRS mode). · EU (2008) standard for protection of vegetation, 16.2 ppb. · Result of cumulative air quality impact assessment: no exceedances.

Final 92 Air Quality Impact Assessment

6.9 Deposition of Gaseous Pollutants

6.9.1 Overview

This section provides a summary of modelling results for the deposition of NH3, NO2 and SO2. The scope of works excludes an environmental impact assessment or analysis of these results as there are no approved deposition standards for the assessment of environmental impacts on land surfaces (for the assessment of vegetation impacts, see Section 6.8).

Results are provided only for the normal operating scenarios expected to run for a whole year; i.e., not for upset condition scenarios, which may last for a period of days or a few weeks only.

Results are provided for deposition of the gaseous pollutants NO2, SO2 and NH3 to enable comparisons with monitoring results described by Gillett (2008), Gillett et al. (2012), and Strategen (2018). These monitoring results are reported as annual amounts, so the results determined from modelling were annual amounts also; provided in units of kilograms per hectare per year (kg/ha/year), and milliequivalents per metres squared per year (meq/m2/year). For brevity, the contour plots provided in this section are provided only in units of meq/m2/year.

Results for NH3 were determined using estimates for the fall velocities of each of the gaseous molecules combined with ground level concentrations. All the results for NH3 deposition were determined using TAPM in tracer mode; see Section 5.5 for methods.

Results for SO2 and NO2 deposition for the Perdaman scenario in isolation are not provided primarily because TAPM does not provide results for deposition in tracer mode. However, the SO2 and NO2 deposition effects due to Perdaman alone can still be seen from comparisons between the baseline results and the other scenarios tested in GRS mode, which included deposition results as outputs.

Final 93 Air Quality Impact Assessment

6.9.2 Scenario PNO – Annual NH3 Deposition

Figure 6-43: PNO – Annual NH3 deposition (meq/m2/year) · Perdaman in isolation – normal operations (PNO); results do not include background. · Maximum grid receptor deposition, 144 meq/m2/year – confined to a small area around the PU Project (the second and third-highest results are 80 and 58 meq/m2/year); median 1.1 meq/m2/year.

Final 94 Air Quality Impact Assessment

6.9.3 Scenario Total Ammonia – Annual NH3 Deposition

Figure 6-44: Total Ammonia – Annual NH3 deposition (meq/m2/year)

· Total Ammonia: Perdaman, Yara Pilbara ammonium nitrate and fertiliser plants, and NH3 background 0.35 µg/m3 (see Section 3.3). · Maximum grid receptor deposition, 148 meq/m2/year (slightly higher than for PNO) and confined to a small area (second and third-highest results are 84 and 62 meq/m2/year); median 5.0 meq/m2/year.

2 · Note from inspection of the contour lines the background NH3 deposition is approximately 4 meq/m /year, which corresponds well with measurements; see Section 3.7.

Final 95 Air Quality Impact Assessment

6.9.4 Scenario Baseline – Annual NO2 deposition

Figure 6-45: Baseline – Annual NO2 deposition (meq/m2/year) · Maximum grid receptor deposition, 12.4 meq/m2/year; median 1.2 meq/m2/year. · No assessment (for information only).

Final 96 Air Quality Impact Assessment

6.9.5 Scenario BPNO – Annual NO2 deposition

Figure 6-46: BPNO – Annual NO2 deposition (meq/m2/year) · Maximum grid receptor deposition, 14.0 meq/m2/year; median 1.2 meq/m2/year. · No assessment (for information only).

Final 97 Air Quality Impact Assessment

6.9.6 Scenario FPNO – Annual NO2 deposition

Figure 6-47: FPNO – Annual NO2 deposition (meq/m2/year) · Maximum grid receptor deposition, 14.9 meq/m2/year; median 1.3 meq/m2/year. · No assessment (for information only).

Final 98 Air Quality Impact Assessment

6.9.7 Scenario Baseline – Annual SO2 deposition

Figure 6-48: Baseline – Annual SO2 deposition (meq/m2/year) · Maximum grid receptor deposition, 42.5 meq/m2/year; median 8.9 meq/m2/year.

· Note SO2 deposition is overestimated near the shipping berths where continuous sources were located in the modelling (conservative, high).

Final 99 Air Quality Impact Assessment

6.9.8 Scenario BPNO – Annual SO2 deposition

Figure 6-49: BPNO – Annual SO2 deposition (meq/m2/year) · Maximum grid receptor deposition, 42.6 meq/m2/year; median 8.9 meq/m2/year.

· Note SO2 deposition is overestimated near the shipping berths where continuous sources were located in the modelling (conservative, high).

Final 100 Air Quality Impact Assessment

6.9.9 Scenario FPNO – Annual SO2 deposition

Figure 6-50: FPNO – Annual SO2 deposition (meq/m2/year) · Maximum grid receptor deposition, 42.6 meq/m2/year; 8.9 meq/m2/year.

· These SO2 deposition results are overestimated near shipping berths where continuous sources were located in the modelling.

Final 101 Air Quality Impact Assessment

6.10 Particulate Urea Deposition

This section provides the TAPM results for particulate urea deposition using the PM10 emissions estimates for the PU Project granulator stacks. TAPM was used in particulate mode, with a PM2.5/PM10 ratio of 30% (see Section 4.5.2).

The TAPM results for particulate urea as PM10 deposition (kg/ha/year), are provided in Figure 6-51.

Figure 6-51: PU Project Particulate Urea Deposition – Annual PM10 deposition (kg/ha/year) · Maximum grid receptor deposition, 9.0 kg/ha/year; median 0.01 kg/ha/year. · Typical high (90th percentile) of 2601 grid receptor results, 0.06 kg/ha/year.

Final 102 Air Quality Impact Assessment

The TAPM results for particulate urea deposition (kg/ha/year), for the PM2.5 component, are provided in Figure 6-52.

Figure 6-52: PU Project Particulate Urea Deposition – Annual PM2.5 deposition (kg/ha/year) · Maximum grid receptor deposition, 0.06 kg/ha/year; median 0.002 kg/ha/year. · Typical high (90th percentile) of 2601 grid receptor results, 0.002 kg/ha/year.

Final 103 Air Quality Impact Assessment

6.11 Summary of Results

6.11.1 Summary of Airborne Results – Grid Receptors

A summary of the TAPM tracer mode and TAPM-GRS results for the (airborne) GLCs for the grid receptor maxima, with corresponding standards for the protection of human health, is provided in Table 6-6.

Table 6-6: Summary of Results: Grid Receptor Maxima and Standards

Assessment Parameter & Air Quality Standard (units) PNO1, 2 PUC1 Baseline BPNO BPUC FPNO [standard]

3 max 24h PM10 (µg/m ) [50] 12.4 19.1 35.5 44.7 53.0 44.6

3 annual PM10 (µg/m ) [25] 6.4 N/A 24.8 30.9 N/A 30.8

3 max 24h PM2.5 (µg/m ) [25] 3.7 5.7 15.5 17.4 18.9 17.4

3 annual PM2.5 (µg/m ) [8] 1.9 N/A 8.4 10.3 N/A 10.3

3 4 max 1h NH3 (µg/m ) [NSW-330] 76.9 75.9 0.4 77.3 76.2 77.3

max 1h NO2 (ppb) [120] 5.1 3.9 42.6 43.1 42.9 43.9

annual NO2 (ppb) [30] 0.5 N/A 5.0 5.6 N/A 5.9

3 3 max 1h O3 (ppb) [100] -- -- 61.8 62.0 61.9 63.0

3 3 max 4h O3 (ppb) [80] -- -- 58.2 58.3 58.2 59.7

max 1h SO2 (ppb) [200] 0.17 0.14 18.2 18.1 18.1 18.1

max 24h SO2 (ppb) [80] 0.05 0.04 7.0 7.0 7.0 7.0

annual SO2 (ppb) [20] 0.02 N/A 4.5 4.5 N/A 4.5

Note 1. TAPM-tracer mode results (PNO and PUC) assume a PM2.5/PM10 ratio of 25% and NO2/NOx ratio of 30%. Note 2. Annual averages for PNO incorporate an annual operating capacity factor of 90.6%. Because of the capacity factor, annual average parameters for BPNO and FPNO are overestimated to a small extent.

Note 3. Results for O3 not available for PNO and PUC (TAPM run in tracer mode).

Note 4. Yara sources had only a small effect on the NH3 GLCs.

A summary of the TAPM-GRS results for (airborne) GLCs for the grid receptor maxima and corresponding standards for the protection of vegetation is provided in Table 6-7.

Table 6-7: Summary of Results: Grid Receptor Maxima and EU 2008 Standards

Assessment Parameter & PNO Baseline BPNO FPNO EU 2008 Standard

annual NOx (ppb); 16.2 ppb 0.3 7.7 8.6 9.1

annual SO2 (ppb); 7.8 ppb 0.02 4.5 4.5 4.5

o o EU 2008 standards: NOX: 16.2 ppb at 30 C; SO2: 7.8 ppb at 30 C.

6.11.2 Summary of Airborne Results – Discrete Receptors Summaries of TAPM tracer mode and TAPM-GRS results for airborne GLCs for Woodside’s Air Quality Monitoring Station (AQMS) locations representing sensitive receptor locations and corresponding standards for the protection of human health, are provided in the following tables for each scenario: Table 6-8 (PNO), Table

Final 104 Air Quality Impact Assessment

6-9 (PUC), Table 6-10 (Baseline), Table 6-11 (BPUC), Table 6-12 (BPNO), Table 6-13 (FPNO). A complete set of results including additional discrete receptor locations is provided in Appendix E.

Table 6-8: Summary of Results PNO: Discrete Receptor Maxima and Standards

Assessment Parameter & Air Quality AQMS Burrup AQMS Dampier AQMS Karratha Grid Maximum* Standard [standard µg/m3] Road

Max 1h NO2 (NO2/NOx 100%) [226] 18.5 5.3 5.9 31.7

Max 1h NO2 (NO2/NOx 30%) [226] 5.5 1.6 1.8 9.5

Annual NO2 (NO2/NOx 100%) [56] 0.51 0.11 0.11 3.3

Annual NO2 (NO2/NOx 30%) [56] 0.15 0.03 0.03 1.0

Annual NO2 incl. NO2/NOx 30% and capacity 0.14 0.03 0.03 0.9 factor 90.6% [56]

Max 1h SO2 200 ppb at 25 °C [524] 0.3 0.1 0.1 0.4

Max 24h SO2 80 ppb at 25 °C [209] 0.07 0.02 0.01 0.12

Annual SO2 20 ppb at 25 °C [52.4] 0.008 0.002 0.002 0.05

Annual SO2 incl. capacity factor 90.6% [52.4] 0.008 0.002 0.001 0.05

Max 24h PM10 [50] 5.6 1.4 1.6 12.4

Annual PM10 [25] 0.80 0.14 0.13 7.0

Annual PM10 0.72 0.12 0.12 6.4 Incl. capacity factor (90.6%) [25]

Max 24h PM2.5 (PM2.5/PM10 ratio = 25%) [25] 1.4 0.4 0.4 3.7

Annual PM2.5 including capacity factor 90.6% 0.2 0.0 0.0 1.9 and PM2.5/PM10 ratio 25% [8]

Max 1h Ammonia (NH3) [330] 33.7 17.0 8.7 76.9

Max 1h Methanol [3000] 0.01 0.00 0.00 0.02

Max 1h CH2O [20] 0.11 0.03 0.03 0.14

Max 24h CH2O [49] 0.02 0.01 0.01 0.04

Note 1. AQMS – ‘Air Quality Monitoring Station’ (Woodside monitoring location). Note 2. TAPM tracer mode results (all units µg/m3). Note 3. All grid maxima at same point near PU Project: E. 476,489 m; N. 7,718,717 m.

Final 105 Air Quality Impact Assessment

Table 6-9: Summary of Results PUC: Discrete Receptor Maxima and Standards

Assessment Parameter & Air Quality AQMS Burrup AQMS Dampier AQMS Karratha Grid Maximum Standard [standard µg/m3] Road

Max 1h NO2 (NO2/NOx 100%) [226] 15.0 4.0 4.5 24.2

Max 1h NO2 (NO2/NOx 30%) [226] 4.5 1.2 1.4 7.2

Max 1h SO2 200 ppb at 25 °C [524] 0.2 0.1 0.1 0.37

Max 24h SO2 80ppb at 25 °C [209] 0.05 0.01 0.01 0.11

Max 24h PM10 [50] 8.2 1.8 2.1 19.1

Max 24h PM2.5 (PM10 ratio = 25%) [25] 2.1 0.4 0.5 5.7

Max 1h Ammonia (NH3) [330] 31.5 16.2 8.7 75.9

Note 1. AQMS – ‘Air Quality Monitoring Station’ (Woodside monitoring location). Note 2. TAPM tracer mode results (all units µg/m3). Note 3. All grid maxima at same point near PU Project: E. 476,489 m; N. 7,718,717 m. Note 4. Annual average results are not applicable to the short-term PUC air emissions scenario.

Final 106 Air Quality Impact Assessment

Table 6-10: Summary of Results Baseline: Discrete Receptor Maxima and Standards

Assessment Parameter & Air Quality AQMS Burrup AQMS Dampier AQMS Karratha Grid Maximum Standard Road

Max 1h NO2 (ppb) [120] 34.5 24.8 24.9 42.6

Annual without CF NO2 (ppb) [30] 3.2 1.7 0.9 5.0

Max 1h O3 (ppb) [100] 58.8 55.4 58.2 61.8

Max 4h O3 (ppb) [80] 54.3 52.6 56.6 58.2

Max 1h SO2 (ppb) [200] 11.2 13.2 3.6 18.2

Max 24h SO2 (ppb) [80] 4.7 4.5 1.7 7.0

Annual SO2 without CF (ppb) [20] 2.0 1.6 0.9 4.5

3 Max 24h PM10 (µg/m ) [25] 34.4 34.5 34.1 35.5

3 Annual PM10 without CF (µg/m ) [25] 23.8 23.7 23.8 24.8

3 Max 24h PM2.5 (µg/m ) [8] 15.0 15.3 14.5 15.5

3 Annual PM2.5 without CF (µg/m ) [8] 8.1 7.9 7.9 8.4

Max 1h Ammonia (NH3) [330] 0.8 0.7 0.9 0.4

Table 6-11: Summary of Results BPUC: Discrete Receptor Maxima and Standards

Assessment Parameter & Air AQMS Burrup Road AQMS Dampier AQMS Karratha Grid Maximum Quality Standard

Max 1h NO2 (NO2/NOx 33.5 24.8 25.4 42.9 100%) (ppb) [120]

Max 1h O3 (ppb) [100] 58.7 55.4 58.4 61.9

Max 4h O3 (ppb) [80] 54.1 52.6 56.7 58.2

Max 1h SO2 (ppb) [200] 11.3 12.9 3.6 18.1

Max 24h SO2 (ppb) [80] 4.7 4.6 1.7 7.0

3 Max 24h PM10 (µg/m ) 37.9 34.7 34.5 53.0 [50]

3 Max 24h PM2.5 (µg/m ) 15.2 15.5 14.8 18.9 [25]

Max 1h Ammonia (NH3) 32.0 16.6 9.1 76.2 (µg/m3) [330]

Final 107 Air Quality Impact Assessment

Table 6-12: Summary of Results BPNO: Discrete Receptor Maxima and Standards

Assessment Parameter & Air AQMS Burrup Road AQMS Dampier AQMS Karratha Grid Maximum Quality Standard

Max 1h NO2 (NO2/NOx 33.6 24.8 25.6 43.1 100%) (ppb) [120]

Annual NO2 (ppb) [30] 3.3 1.7 0.9 5.6

Max 1h O3 (ppb) [100] 58.7 55.4 58.6 62.0

Max 4h O3 (ppb) [80] 54.0 52.7 56.9 58.3

Max 1h SO2 (ppb) [200] 11.3 12.9 3.6 18.1

Max 24h SO2 (ppb) [80] 4.7 4.6 1.7 7.0

Annual SO2 (ppb) [20] 2.0 1.6 0.9 4.5

3 Max 24h PM10 (µg/m ) 36.6 34.6 34.4 44.7 [50]

3 Annual PM10 (µg/m ) 24.5 23.8 23.9 30.9 [25]

3 Max 24h PM2.5 (µg/m ) 15.2 15.5 14.7 17.4 [25]

3 Annual PM2.5 (µg/m ) [8] 8.3 8.0 7.9 10.3

Max 1h Ammonia (NH3) 34.2 17.4 9.1 77.3 (µg/m3) [330]

Final 108 Air Quality Impact Assessment

Table 6-13: Summary of Results FPNO: Discrete Receptor Maxima and Standards

Assessment Parameter & Air AQMS Burrup Road AQMS Dampier AQMS Karratha Grid Maximum Quality Standard

Max 1h NO2 (NO2/NOx 34.2 25.8 28.4 43.9 100%) (ppb) [120]

Annual NO2 (ppb) [30] 4.0 1.8 1.0 5.9

Max 1h O3 (ppb) [100] 58.4 56.5 61.2 63.0

Max 4h O3 (ppb) [80] 53.7 53.6 59.1 59.7

Max 1h SO2 (ppb) [200] 11.4 12.9 3.6 18.1

Max 24h SO2 (ppb) [80] 4.8 4.6 1.7 7.0

Annual SO2 (ppb) [20] 2.0 1.6 0.9 4.5

3 Max 24h PM10 (µg/m ) 36.6 34.7 34.4 44.6 [50]

3 Annual PM10 (µg/m ) 24.6 23.8 23.9 30.8 [25]

3 Max 24h PM2.5 (µg/m ) 15.3 15.5 14.8 17.4 [25]

3 Annual PM2.5 (µg/m ) [8] 8.3 8.0 7.9 10.3

Max 1h Ammonia (NH3) 34.2 17.4 9.1 77.3 (µg/m3) [330]

A summary of the TAPM tracer mode and TAPM-GRS results for the (airborne) GLCs for the discrete receptor maxima, with corresponding standards for the protection of vegetation, is provided in the following tables, one for each scenario: Table 6-14 (PNO), Table 6-15 (Baseline), Table 6-16 (BPNO), Table 6-17 (FPNO). The complete set of discrete receptor results are provided in Appendix F.

Table 6-14: Summary of Results PNO: Discrete Receptor Maxima and EU 2008 Standards

Assessment Parameter AQMS Burrup AQMS Dampier AQMS Karratha Grid maximum [EU 2008 Standard] Road

3 Annual NOx; [30 µg/m ] 0.51 0.11 0.11 3.3

Annual NOx incl. capacity factor 90.6%; 3.0 [30 µg/m3] 0.47 0.10 0.10

3 Annual SO2 [20 µg/m ] 0.008 0.002 0.002 0.05

Annual SO2 incl. capacity factor 90.6% 0.05 [20 µg/m3] 0.008 0.002 0.001

Note: AQMS – ‘Air Quality Monitoring Station’ (Woodside monitoring location); units µg/m3.

Final 109 Air Quality Impact Assessment

Table 6-15: Summary of Results Baseline: Discrete Receptor Maxima and EU 2008 Standards

Assessment Parameter AQMS Burrup AQMS Dampier AQMS Karratha Grid maximum [EU 2008 Standard] Road

3 o Annual NOx (30 µg/m , T = 30 C); 16.2 ppb 4.0 2.1 1.1 7.7

3 o Annual SO2 (20 µg/m , T = 30 C); 7.8 ppb 2.0 1.6 0.9 4.5

Table 6-16: Summary of Results BPNO: Discrete Receptor Maxima and EU 2008 Standards

Assessment Parameter AQMS Burrup AQMS Dampier AQMS Karratha Grid maximum [EU 2008 Standard] Rd

3 o Annual NOx (30 µg/m , T = 30 C); 16.2 ppb 4.2 2.1 1.1 8.6

3 o Annual SO2 (20 µg/m , T = 30 C); 7.8 ppb 2.0 1.6 0.9 4.5

Table 6-17: Summary of Results FPNO: Discrete Receptor Maxima and EU 2008 Standards

Assessment Parameter AQMS Burrup AQMS Dampier AQMS Karratha Grid maximum [EU 2008 Standard] Rd

3 o Annual NOx (30 µg/m , T = 30 C); 16.2 ppb 5.2 2.3 1.3 9.0

3 o Annual SO2 (20 µg/m , T = 30 C); 7.8 ppb 2.0 1.6 0.9 4.5

6.11.3 Summary of Deposition Results – Grid Receptors

A summary of the TAPM (tracer mode) TAPM-GRS (photochemical mode) results for annual deposition is provided in Table 6-18. The two shorter-term upset conditions scenarios are excluded from these annual statistics. The medians of the 2,601 GR results are provided as indicators of typical deposition values predicted by modelling and the calculations; the maxima were confined to relatively small areas around the PU Project site (see contour plots provided in Section 6.9). The results are compared with measurements in Section 7.

Table 6-18: Summary of Results for Deposition: Grid Receptor Maxima and Medians

Annual Deposition PNO* Baseline BPNO FBNO Total NH3

NH3 (kg/ha/year) 24.6, 0.2 See Total NH3* See Total NH3 See Total NH3 25.3, 0.8

NH3 (kg/ha/year) with 22.3, 0.2 - - - 22.9, 0.8 capacity factor 90.6%

NO2 (kg/ha/year) # 5.7, 0.5 6.4, 0.5 6.8, 0.6 #

SO2 (kg/ha/year) # 13.6, 2.8 13.7, 2.8 13.7, 2.8 #

2 NH3 (meq/m /year) 144.3, 1.1 See Total NH3 See Total NH3 See Total NH3 148.3, 5.0

2 NH3 (meq/m /year) with 130.7, 1.0 - - - 134.3, 4.5 capacity factor 90.6%

2 NO2 (meq/m /year) # 12.4, 1.2 14.0, 1.2 14.9, 1.3 #

2 SO2 (meq/m /year) # 42.5, 8.9 42.6, 8.9 42.6, 8.9 #

*NH3 deposition calculated using TAPM tracer-mode results.

# NO2 and SO2 deposition from outputs from TAPM-GRS.

Final 110 Air Quality Impact Assessment

6.11.4 Summary of Deposition Results – Discrete Receptors

Summaries of TAPM tracer mode and TAPM-GRS results for deposition for Woodside’s Air Quality monitoring Station (AQMS) locations representing sensitive receptor locations are provided in the following tables for each scenario.

Table 6-19: Comparisons of Model Results: NH3 Deposition (meq/m2/year)

Monitoring Total NH3 Total NH3 PNO Baseline Easting (m) Northing (m) Station BPNO FPNO

Dolphin Island 1.7 3.9 5.6 5.6 484,598 7,738,456

North Burrup 3.0 4.0 7.0 7.0 482,347 7,730,288

Woodside East 10.9 4.2 15.1 15.1 477,363 7,721,921

Burrup Road 32.1 4.2 36.3 36.3 475,961 7,719,787

Water Tank 21.2 5.6 26.8 26.8 477,616 7,720,114

Ngajarli* 37.1 4.0 41.1 41.1 477,964 7,718,020

King Bay South 3.5 3.9 7.4 7.4 474,026 7,717,213

Karratha 1.4 3.9 5.4 5.4 482,990 7,707,089

Mardie Station 0.0 3.9 3.9 3.9 408,643 7,659,017

Hearson Cove 39.2 3.9 43.0 43.0 478,928 7,718,358

MNP–CN 2.9 4.0 6.9 6.9 483,354 7,730,501

MNP–CS 3.9 3.9 7.8 7.8 476,195 7,714,869

King Bay, MAC 31.6 4.2 35.7 35.7 475,574 7,719,459

Standing Stones 7.4 3.9 11.3 11.3 474,714 7,717,782

MLKC 4.4 4.1 8.4 8.4 479,900 7,727,100

Names: *’Ngajarli’, formerly ‘Deep Gorge’. MNP–CN: Murujuga National Park – Central Northern representative. MNP–CS: Murujuga National Park – Central Southern representative. MAC: Murujuga Aboriginal Corporation. MLKC: Murujuga Living Knowledge Centre.

Methods: (1) Discrete receptor list based on Gillett (2008).

(2) Baseline determined by difference between Total-NH3 and PNO scenarios. (3) FPNO equivalent to BPNO. (4) Mardie results estimated by inspection of data and plots.

(5) Deposition for NH3 calculated from TAPM results; tracer mode.

Final 111 Air Quality Impact Assessment

Table 6-20: Comparisons of Model Results: NO2 Deposition (meq/m2/year)

Monitoring PNO Baseline BPNO FPNO Easting (m) Northing (m) Station

Dolphin Island 0.1 1.6 1.7 1.8 484,598 7,738,456

North Burrup 0.1 2.1 2.2 2.5 482,347 7,730,288

Woodside East 0.3 6.2 6.5 7.4 477,363 7,721,921

Burrup Road 0.6 6.5 7.1 9.4 475,961 7,719,787

Water Tank 0.4 8.1 8.6 10.8 477,616 7,720,114

Ngajarli 1.6 9.1 10.7 11.9 477,964 7,718,020

King Bay South 0.2 4.3 4.4 4.8 474,026 7,717,213

Karratha 0.1 1.8 1.9 2.1 482,990 7,707,089

Mardie Station 0.0 1.0 1.0 1.0 408,643 7,659,017

Hearson Cove 1.0 6.9 7.9 9.0 478,928 7,718,358

MNP–CN 0.1 2.1 2.2 2.6 483,354 7,730,501

MNP–CS 0.2 3.0 3.2 3.6 476,195 7,714,869

King Bay, MAC 0.7 6.4 7.1 9.0 475,574 7,719,459

Standing Stones 0.3 5.4 5.7 6.1 474,714 7,717,782

MLKC 0.2 2.9 3.1 3.5 479,900 7,727,100

Names: *’Ngajarli’, formerly ‘Deep Gorge’. MNP–CN: Murujuga National Park – Central Northern representative. MNP–CS: Murujuga National Park – Central Southern representative. MAC: Murujuga Aboriginal Corporation. MLKC: Murujuga Living Knowledge Centre.

Methods: (1) Discrete receptor list based on Gillett (2008). (2) Mardie estimated by inspection of plots.

(3) NO2 deposition by TAPM in GRS mode. (4) PNO results by differences BPNO minus background (background estimated by inspection of plots).

Final 112 Air Quality Impact Assessment

Table 6-21: Comparisons of Model Results: SO2 Deposition (meq/m2/year)

Monitoring PNO Baseline BPNO FPNO Easting (m) Northing (m) Station

Dolphin Island 0.0 7.0 7.0 7.1 484,598 7,738,456

North Burrup 0.0 5.9 5.9 5.9 482,347 7,730,288

Woodside East 0.0 6.4 6.4 6.4 477,363 7,721,921

Burrup Road 0.0 6.4 6.5 6.5 475,961 7,719,787

Water Tank 0.0 6.1 6.1 6.2 477,616 7,720,114

Ngajarli 0.0 6.8 6.8 6.8 477,964 7,718,020

King Bay South 0.0 10.0 10.0 10.0 474,026 7,717,213

Karratha 0.0 2.4 2.4 2.4 482,990 7,707,089

Mardie Station 0.0 2.0 2.0 2.0 408,643 7,659,017

Hearson Cove 0.1 9.1 9.2 9.2 478,928 7,718,358

MNP–CN 0.0 6.1 6.1 6.1 483,354 7,730,501

MNP–CS 0.0 6.4 6.4 6.4 476,195 7,714,869

King Bay, MAC 0.0 6.0 6.1 6.1 475,574 7,719,459

Standing Stones 0.0 9.2 9.2 9.2 474,714 7,717,782

MLKC 0.0 6.0 6.0 6.0 479,900 7,727,100

Names: *’Ngajarli’, formerly ‘Deep Gorge’. MNP–CN: Murujuga National Park – Central Northern representative. MNP–CS: Murujuga National Park – Central Southern representative. MAC: Murujuga Aboriginal Corporation. MLKC: Murujuga Living Knowledge Centre.

Methods: (1) Discrete receptor list based on Gillett (2008). (2) Mardie estimated by inspection of plots.

(3) SO2 deposition by TAPM in GRS mode. (4) PNO results by differences BPNO minus background (background estimated by inspection of plots).

Final 113 Air Quality Impact Assessment

7. Comparisons with Deposition Monitoring

7.1.1 Overview

The purpose of this section is to compare the results of modelling and calculations for deposition with measurements from two monitoring programs conducted over 2004-2005 and 2007-2008, and some updates from Woodside (2019); see Section 3.7. The monitoring results in each measurement phase were similar, so only the more recent 2007-2008 dataset was used for the detailed comparisons in this section.

The Mardie Station used to determine ‘background’ or baseline deposition is off the study grid (off the TAPM inner-grid to the south-west), so model results for that location were estimated by inspection of the numerical results and the contour plots.

7.1.2 Summary of Dry Deposition Results – Monitoring Locations

The results of modelling and calculations for NH3 dry deposition were estimated by interpolation for the ten monitoring locations described in Gillett (2008) and Gillett et al. (2012). Baseline modelled NH3 results were determined by the differences between the results for the PNO and Total NH3 scenarios. Results for dry deposition of NH3 are provided for comparison in Figure 7-1.

Figure 7-1: NH3 Dry Deposition: Modelling and Monitoring (meq/m2/year)

The comparisons show that the main discrepancy between the modelled baseline and monitoring of NH3 dry deposition occurs at Karratha; this may have been due to cattle feedlots or the use of fertiliser in the Karratha area not accounted for in the modelling (see Section 3.7). For the other locations, the modelling is underestimating NH3 dry deposition – again, this may be due to the omission of (unknown) agricultural sources in the modelling.

Final 114 Air Quality Impact Assessment

Results for dry deposition of NO2 are provided for comparison in Figure 7-2. The monitoring results are compared with TAPM-GRS results for the current existing air emissions scenario (Baseline), and the proposed future Baseline including emissions from Perdaman. In general, there is good agreement between the results. While the modelling has overestimated NO2 for the Water Tank and Ngajarli (Deep Gorge) locations, an increase in industrial activity on the Burrup Peninsula since the measurements of 2007-2008 means these model results may be providing a more accurate indication of NO2 dry deposition than the older monitoring results at those locations. The reason for the model underestimating NO2 deposition in Karratha is unknown but would be due to a local NO2 source not included in the modelling – possibly vehicle traffic. In any case there are no receptors in Karratha sensitive to these small NO2 deposition amounts. There is only a small difference between the results for the distant background site Mardie Station, indicating the model predicted background levels correctly.

Figure 7-2: NO2 Dry Deposition: Modelling and Monitoring (meq/m2/year)

Comparisons of results for SO2 dry deposition are provided in Figure 7-3. In this case, even though the monitoring results are from over a decade ago, the comparisons indicate the model results are overestimating SO2 dry deposition around the Burrup Peninsula. This is not unexpected, as the SO2 emissions from shipping were modelled conservatively by assuming continuous emissions from ship exhausts at every berth on the Burrup Peninsula.

A final assessment of the accuracy of the deposition results was undertaken by comparing the summations of the modelled dry depositions with measurements. The results are provided for comparison in Figure 7-4. This shows the summation of YPN (2017) measurements for HNO3, NH3, NO2 and SO2, and the summation of model results for NH3, NO2 and SO2 (no model results for HNO3). Summaries of more recent measurements indicate typical values for total N and S dry deposition on Burrup Peninsula are in the range 20–30 meq/m2/year (see Section 3.7; Gillett, 2014; Woodside, 2019).

Final 115 Air Quality Impact Assessment

Figure 7-3: SO2 Dry Deposition: Modelling and Monitoring (meq/m2/year)

Figure 7-4: Summations of N and S Dry Deposition: Modelling and Monitoring (meq/m2/year)

Final 116 Air Quality Impact Assessment

8. Comparisons with Air Pollutant Monitoring

The purpose of this section is to compare key statistical results from the current TAPM-GRS modelling with corresponding statistics from the 2014 monitoring results–2014 was selected as the simulated meteorological year for modelling; see Section 5.3.5.

Comparisons of the TAPM results for hourly average NO2 GLCs (ppb) with monitoring data are set out in Table 8-1. The plots provide statistical summaries of the 8760 one-hour average NO2 GLCs predicted by TAPM-GRS for three grid point locations representative of the Karratha (left), Dampier (middle) and Burrup Road (right) monitoring locations. The TAPM ‘CLOC’ parameter captures the maximum grid point concentration surrounding the selected point, so provides a better indication of the broader model results for each location.

A similar comparison of modelling vs. monitoring results for the 2014 year is provided in Table 8-2 for O3. In 2014, O3 monitoring data were obtained from Karratha and Dampier monitoring stations only.

The Robust Highest Concentration (RHC) is an estimate of the maximum, which attempts to minimise over- estimates or under-estimates in a dataset; e.g., see Hurley (2008a). Estimates for the RHCs are also provided in the following tables. The hourly average statistics plotted (left-to-right) in each chart are: maximum, RHC, 99.9th percentile, 99th percentile, 70th percentile, 50th percentile (i.e. median), and annual average. An analysis of the comparisons is provided below each chart.

The reliability of the TAPM-GRS results was determined primarily by comparisons of model results with monitoring records. These comparisons of statistical results indicated TAPM-GRS was performing well in terms of being able to accurately predict a variety of statistical results for NO2 and O3 as measured by Woodside at the Burrup, Dampier and Karratha monitoring stations.

Final 117 Air Quality Impact Assessment

Table 8-1: Comparisons of TAPM Results with 2014 Monitoring Results for Hourly Average NO2

· Karratha 2014: 1-Hour Average NO2 (ppb) · Dampier 2014: 1-Hour Average NO2 (ppb) · Burrup 2014: 1-Hour Average NO2 (ppb) · TAPM (blue) and monitoring (yellow) · TAPM (blue) and monitoring (yellow) · TAPM (blue) and monitoring (yellow) · Plotted range is 0-60 ppb · Plotted range is 0-60 ppb · Plotted range is 0-60 ppb · NEPM (Ambient Air Quality) standard 120 · NEPM (Ambient Air Quality) standard 120 · NEPM (Ambient Air Quality) standard 120 ppb ppb ppb Analysis: Analysis: Analysis:

Generally good agreement between the TAPM Excellent agreement between the TAPM results Excellent agreement between the TAPM results results and monitoring for the higher NO2 and monitoring for the higher NO2 concentrations (blue) and monitoring (yellow) for the higher NO2 concentrations in Karratha; e.g., the 99.9th in Dampier. concentrations for Burrup Road; parameter percentile for the grid point selected to represent ‘CLOC’ indicates the TAPM results are Karratha is almost an exact match. CLOC parameter indicates the TAPM results are conservative, high). conservative, high. TAPM slightly underestimating annual average Good agreement for annual average NO2 at NO2 for both point ‘Karratha’ and ‘CLOC’. Excellent agreement for annual average NO2 at Dampier, with TAPM overestimating Dampier, and TAPM slightly overestimating (conservative, high). (conservative, high).

Final 118 Air Quality Impact Assessment

Table 8-2: Comparisons of TAPM Results with 2014 Monitoring Results for Hourly Average O3

· Karratha 2014: 1-Hour Average O3 (ppb) · Dampier 2014: 1-Hour Average O3 (ppb) · TAPM (blue) and monitoring (yellow) · TAPM (blue) and monitoring (yellow) · Plotted range is 0-100 ppb · Plotted range is 0-100 ppb · NEPM (Ambient Air Quality) standard 100 ppb · NEPM (Ambient Air Quality) standard 100 ppb Excellent overall agreement between the TAPM results for hourly average Excellent overall agreement between the TAPM results for hourly average O3 concentrations and monitoring across the whole range of statistics. The O3 concentrations and monitoring across the whole range of statistics. The comparisons of RHCs is perfect, with TAPM slightly conservative (slightly comparisons of RHCs is perfect, with TAPM slightly conservative (slightly higher). higher).

TAPM overestimating annual average O3 (conservative). TAPM overestimating annual average O3 (conservative).

Final 119 Air Quality Impact Assessment

9. Conclusion

This report details the results of air quality modelling to accompany the environmental approvals for the PU Project. As a part of this assessment the existing air emissions scenario and potential future air emissions scenarios were developed for the Burrup Peninsula. The results of modelling were set out to determine how current emissions affect existing air quality. Potential future air emissions scenarios were modelled to increase our understanding of how the PU Project would affect the current air quality situation.

The CSIRO meteorological, air dispersion and photochemical model, ‘TAPM-GRS’ was selected for modelling for reasons of reliability and efficiency. To confirm TAPM-GRS performance was ‘fit-for-purpose’, modelled results were compared to measured results from Woodside ambient air monitoring programs. TAPM was used in the simpler ‘tracer’ (mass dispersion) mode for the assessment of emissions of ammonia (NH3), formaldehyde, and methanol from the PU Project. The assessment included modelling of particulate matter as PM10 and PM2.5, nitrogen dioxide (NO2), ozone (O3) and sulfur dioxide (SO2) for assessment against standards set out in the National Environmental Protection (Ambient Air Quality) Measure (‘NEPM’).

Model results for airborne concentrations of oxides of nitrogen (NOx) and SO2 were obtained for comparison with European Union (2008) air quality standards for the protection of vegetation.

Model results for NH3, NO2 and SO2 deposition were provided to support any future assessment of potential impacts to landforms, including Murujuga rock art on the Burrup Peninsula.

Also, TAPM was used in particulate mode (PM2.5 and PM10), with the conservative assumption that 100% of the PM10 was assumed to be particulate urea. These emissions were from the PU Project’s granulator stacks, using an assumed PM2.5/PM10 ratio of 30%.

Key results from the air quality impact assessment for the PU Project were:

· TAPM-GRS results for airborne NO2 and O3 compared very well with measurements indicating strongly the model was fit-for-purpose.

· There were no predicted exceedances of NEPM standards for NO2, O3, and SO2 concentrations for any of the emission scenarios that were investigated as part of this assessment. All results for these pollutants were well below relevant NEPM standards.

· There were no predicted exceedances of ambient air quality assessment criteria for NH3, formaldehyde and methanol. (Formaldehyde and methanol were eliminated from detailed assessment as low risk substances).

· There were no predicted exceedances of European Union (2008) air quality standards for NOx and SO2 for the protection of vegetation, for any of the emission scenarios.

· Estimates for dry deposition of gaseous NH3, NO2 and SO2 were determined from a combination of modelled results and calculations, which compared reasonably well with monitoring results for total dry deposition of nitrogen and sulfur-containing gases, adding further weight that the methods applied were fit-for-purpose.

· Examples of results for particulate urea deposition were: as PM10, 0.06 kg/ha/year; and the PM2.5 deposition, 0.002 kg/ha/year (90th percentiles of annual averages for 2601 grid receptor results).

In conclusion, based on modelling which showed compliance with relevant air quality criteria, there is a low risk of air quality impact on human health and vegetation from the PU Project. Model results for airborne concentrations of air pollutants compared well with monitoring. Also, model results for gaseous dry deposition compared reasonably well with monitoring, or typically within 5 meq/m2/year of monitoring results at eight sensitive receptor locations.

Final 120 Air Quality Impact Assessment

References

Air Assessments, 2010a, Browse LNG Development, Air Quality Assessment, Woodside Energy Ltd., Prepared by Air Assessments. Available from: https://www.jtsi.wa.gov.au/docs/default-source/default-document- library/browse_sar_appendix_c-25_1210.pdf?sfvrsn=d2686b1c_10

Air Assessments, 2010b, 2009 Burrup Ambient Air Quality Study, Woodside Energy Ltd., Prepared by Air Assessments.

Air Assessments, 2012, Gorgon Gas Development Fourth Train Proposal, Air Quality Assessment, Chevron Australia, June 2012, Final.

Australian Maritime Safety Authority Australian Maritime Safety Authority (AMSA), 2020, Low sulphur fuel. Available from: https://www.amsa.gov.au/marine-environment/air-pollution/2020-low-sulphur-fuel.

Azzi, M, Johnson, G, Cope, M, 1992 An introduction to the generic reaction set photochemical smog mechanism. Proceedings of the 11th International Clean Air and Environment Conference, Clean Air Society of Australia and New Zealand, Brisbane.

Bureau of Meteorology (BOM), 2019a, Tropical Cyclones Affecting Pilbara, http://www.bom.gov.au/cyclone/history/wa/pilbara.shtml, accessed 28 Feb 2019.

Bureau of Meteorology (BOM), 2019b, Tropical Cyclone Monty, http://www.bom.gov.au/cyclone/history/wa/monty.shtml, accessed 28 Feb 2019.

Burrup Nitrates Pty Ltd, 2009a, Annex A: DEWHA and OEPA Level of Assessment and Guidance, Technical Ammonium Nitrate Production Facility Public Environmental Review Document.

Burrup Nitrates Pty Ltd, 2009b, Annex N: Technical Ammonium Nitrate Plant Air Quality Assessment, Technical Ammonium Nitrate Production Facility Public Environmental Review Document, Prepared for Burrup Nitrates Pty Ltd by Environmental Resource Management Australia 2009.

Cardno, 2019, Perdaman Urea Project Environmental Scoping Document Rev 5.0, EPA Assessment No. 2184 - EPBC 2018/8383, Prepared for Perdaman Chemicals and Fertilisers Pty Ltd., 19 July 2019.

Cope, M, & Lee, S, 2009, Chemical Transport Model User Manual, The Centre for Australian Weather and Climate Research, A partnership between CSIRO and the Bureau of Meteorology.

Department of Environment (DoE), 2004, Air Quality Management Branch, Pilbara Air Quality Study Summary Report, Technical Series, Report No. 120.

Department of Water and Environmental Regulation (DWER), 2019, Murujuga Rock Art Strategy, https://www.der.wa.gov.au/images/documents/our-work/programs/burrup/Murujuga_Rock_Art_Strategy.pdf, accessed 7 Aug 2019.

Environmental Alliances, 2007, Dust Dispersion Modelling for Pilbara Iron Dampier Port Expansion to 145 Mtpa (Phase B), Prepared for Sinclair Knight Merz by Environmental Alliances Pty Ltd, May 2007.

EPA, 2001, Western Australia Environmental Protection Agency, Ammonia Plant, Burrup Peninsula: Report and recommendations of the Environmental Protection Authority. Available from: http://epa.wa.gov.au/sites/default/files/EPA_Report/998_B1036.pdfEPA, 2011, Western Australia Environmental Protection Agency, Technical Ammonium Nitrate Production Facility, Burrup Peninsula: Report and recommendations of the Environmental Protection Authority. Available from: http://www.epa.wa.gov.au/sites/default/files/EPA_Report/Final%20EPA%20Report%20050111-web_0.pdf

Final 121 Air Quality Impact Assessment

EPA, 2016, Western Australia Environmental Protection Agency, Environmental Factor Guideline: Air Quality. Available from: http://www.epa.wa.gov.au/sites/default/files/Policies_and_Guidance/Guideline-Air-Quality- 131216_3.pdf

European Union, 2008, DIRECTIVE 2008/50/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 21 May 2008 on ambient air quality and cleaner air for Europe. Available from: https://www.eea.europa.eu/themes/air/air-quality-standards

GHD, 2009, GHD, Perdaman Chemicals and Fertilisers, Report for Collie Urea Project Air Quality Assessment, September 2009.

Gillett, R, 2008. Burrup Peninsula Air Pollution Study: Report for 2004/2005 and 2007/2008. Report of 10th September by Rob Gillett, CSIRO Marine and Atmospheric Research.

Gillett, R, 2014, Woodside Nitrogen Deposition Study Final Report, abstract downloaded from CSIRO Research Publications Repository (full report is Woodside-Confidential); CSIRO Oceans and Atmosphere Flagship, 10 October 2014.

Gillett, R, Selleck, P, & Powell, J, 2012, An acid deposition study in a remote area of north western Australia, Air Quality and Climate Change, Vol. 46, No. 3, pp.28-34, August 2012.

GWA, 2015, Government of Western Australia, Department of Water and Environmental Regulation, Licence Number L7997/2002/11, Licence Holder Yara Pilbara Fertilisers Pty Ltd (ACN 095 441 151), File Number DER2014/001374, Duration 21 April 2015 to 20 April 2020, Date of issue16 April 2015.

GWA, 2018, Government of Western Australia, Department of Water and Environmental Regulation, Licence Number L7997/2002/11, Licence Holder Yara Pilbara Fertilisers Pty Ltd (ACN 095 441 151) and Yara Pilbara Nitrates Pty Ltd (ACN 127 391 442), File Number DER2013/001374, Duration 21 April 2015 to 20 April 2020, Date of amendment 29 June 2018.

GWA, 2020, Government of Western Australia, Media Statements, Murujuga World Heritage listing one step closer, 28 January 2020, https://www.mediastatements.wa.gov.au/Pages/McGowan/2020/01/Murujuga-World- Heritage-listing-one-step-closer.aspx, accessed 5 March 2020.

Hacker, J, Chen, D, Bai, M, Ewenz, C, Junkermann, W, Lieff, W, McManus, B, Neininger, B, Sun, J, Coates, T, Denmead, T, Flesch, T, McGinn, S & Hill, J, 2016, Using airborne technology to quantify and apportion emissions of CH4 and NH3 from feedlots. Animal Production Science, 56, 190-203, doi:10.1071/AN15513, http://dx.doi.org/10.1071/AN15513, CSIRO Publishing, 2016.

Hurley, P, Physick, W & Cope, M, 2004, Summary of TAPM Verification for the Pilbara Region, A Report to the Department of Environment, WA CSIRO Atmospheric Research. Available from: http://www.cmar.csiro.au/e- print/open/hurley_2004a.pdf

Hurley, P, 2008, TAPM V4. User Manual, CSIRO Marine and Atmospheric Research Internal Report No. 5. Available from: https://www.cmar.csiro.au/research/tapm/docs/tapm_v4_user_manual.pdf

Hurley, P, 2008a, TAPM V4. Part 1: Technical Description, CSIRO Marine and Atmospheric Research, Paper No. 25. Available from: https://publications.csiro.au/rpr/pub?pid=procite:0cff4149-4feb-4b86-abcb-c707168ecb0b

Hurley, P, Edwards, M & Luhar, A, 2008, TAPM V4. Part 2: Summary of Some Verification Studies, CSIRO Marine and Atmospheric Research Paper No. 26. Available from: https://www.cmar.csiro.au/research/tapm/docs/tapm_v4_technical_paper_part2.pdf

Hurley, P, Edwards, M & Luhar, A, 2009, Evaluation of TAPM V4 for Several Meteorological and Air Pollution Datasets, CSIRO Marine and Atmospheric Research, Air Quality & Climate Change, 43-3, 19-24.

Final 122 Air Quality Impact Assessment

Integrated Heritage Services, 2018. Ethnographic Surveys and Audit for Woodside Karratha Gas Plant, Murujuga (Burrup Peninsula), Western Australia.

Jacobs, 2019a, Provision of Environmental Services to Provide Air Quality Modelling to Support Burrup Hub Environmental Approvals – 5A.1 Burrup Hub Strategic Modelling Scope of Work, CTR Reference P18-27a.

Jacobs, 2019b, EPA meeting 24/1/19 - Jacobs meeting notes, Memo from Matthew Pickett to Michael Hanlin (WEL), Cc. Michael Bell, Shane Lakmaker, Sarah Kelly, 25 January 2019.

Jacobs, 2019c, Air Quality Impact Assessment, Pluto LNG Expansion, Woodside Energy Ltd., Revision 1, 28 June 2019.

Jacobs, 2019d, Air Quality Impact Assessment, NWS Project Extension, Woodside Energy Ltd., Revision 0, 4 July 2019.

Keywood, M, Hibberd, M and Emmerson, K, 2017, Australia State of the Environment 2016: Atmosphere, independent report to the Australian Government, Minister for the Environment and Energy, Australian Government Department of the Environment and Energy, Canberra, doi:10.4226/94/58b65c70bc372. MAC, 2016, Murujuga Cultural Management Plan. Available Cultural Management Plan. Available from: https://trove.nla.gov.au/work/218070893?selectedversion=NBD59378542

Mokhatab, S, Poe, W, Mak, J, 2015, Chapter 3 - Basic Concepts of Natural Gas Processing, Handbook of Natural Gas Transmission and Processing (Third Edition), Gulf Professional Publishing, pp. 123-135, ISBN 9780128014998, https://doi.org/10.1016/B978-0-12-801499-8.00003-1.

National Environment Protection Council (NEPC), 2011, National Environment Protection (Air Toxics) Measure. Available from: https://www.legislation.gov.au/Series/F2007B01121

National Environment Protection Council (NEPC), 2016, National Environment Protection (Ambient Air Quality) Measure. Available from: https://www.legislation.gov.au/Series/F2007B01142

NSW Environment Protection Authority (NWS EPA), 2016, Approved Methods for the Modelling and Assessment of Air Pollutants in New South Wales. Available from: https://www.epa.nsw.gov.au/~/media/EPA/Corporate%20Site/resources/epa/approved-methods-for-modelling- and-assessment-of-air-pollutants-in-NSW-160666.ashx

Physick, W, Blockley, A, Farrar, D, Rayner, K & Mountford, P, 2002, Application of Three Air Quality Models to the Pilbara Region, Proceedings of the 16th International Conference of the Clean Air Society of Australia and New Zealand, Christchurch, New Zealand.

Physick et al. (2004): Physick, W, Rayner, K, Mountford, P & Edwards, M, 2004, Observations and modelling of dispersion meteorology in the Pilbara region, Aust. Met. Mag. 53, pp.175-187.

Pilbara Iron, 2019, Dust Monitoring. Available from: http://vdv.benchmarkmonitoring.com.au/vdv/vdv_gmap.php

Rio Tinto, 2015, Frequently asked questions Air quality monitoring, RTIO-HSE-0242270.

Schrader, F & Brümmer, C, Land Use Specific Ammonia Deposition Velocities: a Review of Recent Studies (2004–2013), Water Air Soil Pollution 225:2114, DOI 10.1007/s11270-014-2114-7, Published online: 13 September 2014.

Seinfeld, J & Pandis, S, 2016, Atmospheric Chemistry and Physics: From Air Pollution to Climate Change, 3rd Edition, ISBN: 978-1-118-94740-1.

Final 123 Air Quality Impact Assessment

Shen, J, Chen, D, Bai, M, Sun, J, Coates, T, Lam, S. K., & Li, Y, 2016, Ammonia deposition in the neighbourhood of an intensive cattle feedlot in Victoria, Australia. Scientific Reports, 6(1), 32793-32793. doi:10.1038/srep32793 Nature, 7 September 2016.

SKM, 2009, Burrup Rock Art: Revised Modelling Taking into Account Recent Monitoring Results, Final 1.

Strategen, 2018, Strategen Environmental, Ambient air quality report 2017-2018, EPBC 2008/45546, Prepared for Yara Pilbara Nitrates, October 2018.

Venkatram A, Karamchandani P, Prasad P, Sloane C, Saxena P, and Goldstein R, 1997, The development of a model to examine source-receptor relationships for visibility on the Colorado Plateau, Journal of the Air and Waste Management Association, 47, 286-301.

Woodside, 2019, Woodside, North West Shelf Project Extension, Environmental Review Document, EPA Assessment No. 2186, EPBC 2018/8335, Revision 1 – December 2019.

World Health Organization (WHO), Regional Office for Europe, 2000, Air quality guidelines for Europe, 2nd ed. Copenhagen: WHO Regional Office for Europe. Available from: http://www.euro.who.int/__data/assets/pdf_file/0005/74732/E71922.pdf United States Environmental Protection Agency (US EPA), 1992, Guidelines for Exposure Assessment. Available from: https://www.epa.gov/sites/production/files/2014-11/documents/guidelines_exp_assessment.pdf.

Yara, 2017, Yara Pilbara Nitrates, EPBC Approval 2008/4546 Baseline Air Quality Monitoring Report, Issued 16 June 2017.

YPN, 2017, Yara Pilbara Nitrates, Letter to Environment and Communications References Committee, Re: Results of Baseline Air Quality Monitoring Status on the Burrup Peninsula, 4 October 2017, and Strategen Environmental, Revised briefing note for YPN, Deposition rates from Baseline study Yara Pilbara Technical Ammonium Nitrate Plant, 3 October 2017.

Final 124 Air Quality Impact Assessment

Appendix A. Location Maps of Monitoring Stations

Figure 9-1: Bureau of Meteorology and Air Quality Monitoring Stations

125 Air Quality Impact Assessment

Figure 9-2: Yara Pilbara Nitrates Air Quality Monitoring Stations (Yara, 2017)

126 Air Quality Impact Assessment

Appendix B. Local Meteorology

Overview

Local meteorology is a critical input for determining the direction and rate at which emissions from a source are likely to disperse, near ground level. This section provides climatological summaries of meteorological parameters representative of the Burrup Peninsula based on Bureau of Meteorology (BoM) observations. The closest BoM weather station to the PU Project site is Karratha Aerodrome (BoM station number 004083, 20.71° S, 116.77° E, elevation 5.3m), which is located approximately 12 km south of the PU Project. The following sub- sections provide summaries of meteorological data acquired over more than two decades at Karratha Aerodrome.

B.1 Temperature

Monthly mean maximum and minimum temperatures for BoM Karratha Aerodrome for 1993-2018 are shown in Figure B- 1. Daily maximum and minimum temperatures have ranged from 48oC in the wet season to only 7oC in the dry season, from 1993 to 2018.

Figure B- 1: Monthly Mean-Maximum and Minimum Temperature – Karratha Aerodrome 1993-2018

B.2 Rainfall and Relative Humidity

Monthly rainfall statistics for BoM Karratha Aerodrome are shown in Figure B- 2, and monthly mean 9am and 3pm Relative Humidity (RH) for Karratha Aerodrome for 1993-2010 are shown in Figure B- 3. The rainfall observations clearly show the Burrup Peninsula wet season running from approximately January to June, and the dry season from approximately July to December.

127 Air Quality Impact Assessment

Figure B- 2: Monthly Rainfall – Karratha Aerodrome 1972-2018

Figure B- 3: Monthly 9am and 3pm Relative Humidity – Karratha Aerodrome 1972-2018

128 Air Quality Impact Assessment

B.3 Wind Speed and Wind Patterns

Monthly mean daily wind speeds and maximum wind gusts for BoM Karratha Aerodrome for 2003-2018 are shown in Figure B- 4.

Figure B- 4: Mean Daily Wind Speed and Maximum Wind Gust – Karratha Aerodrome 1993-2018

The 2014 examples are shown in Figure B- 5. The wind roses show westerly winds were dominant during summer and spring over 2010-2018. There was significantly more annual variability in the wind patterns for autumn and winter (see Figure B- 4), but this may be an artefact of the artificial boundaries of those seasons in relation to the Pilbara’s dry and wet seasons.

Hourly average wind speed statistics calculated from measurements at BoM Karratha and two other weather stations in the Burrup region in 2014, are compared in Table B- 1. The wind speeds at Karratha match those of Roebourne reasonably well. Higher wind speeds were observed at the more exposed site at Legendre Island just north of the peninsula.

Table B- 1: Wind Speed Comparisons – Burrup Peninsula 2014

Statistic BoM Karratha Aerodrome BoM Roebourne BoM Legendre Island

Data Capture % 99.9% 99.9% 99.9%

Maximum (m/s) 13.1 13.4 16.1

90th percentile (m/s) 8.0 7.8 9.7

70th percentile (m/s) 6.2 5.7 7.1

Average (m/s) 5.0 4.5 6.0

129 Air Quality Impact Assessment

Figure B- 5: Annual and Seasonal Wind Roses for 2014 – BoM Karratha Aerodrome*

A full set of BoM Karratha Aerodrome wind roses for 2010-2018 is provided in the final section of this Appendix.

130 Air Quality Impact Assessment

B.4 Pilbara Cyclones

Cyclones have affected the coastal communities of Port Hedland, Karratha, Dampier, and Onslow, and parts of inland Pilbara. Typically, these cyclones form over warm ocean waters to the north, intensify before crossing the Pilbara coast, then track towards the south. The further south they move the more likely they will move south- easterly across inland parts of WA (BoM, 2019a). For example, the track of Tropical Cyclone Monty, 27 February to 2 March 2004, is shown in Figure B- 6 (BoM, 2019b).

Figure B- 6: Track of Tropical Cyclone Monty 2004 (BoM, 2019b)

Heavy rainfall and flooding are the main impacts for most cyclonic events in inland Pilbara. The highest rainfall is usually found along or just east of the track for most systems. The flood potential of a cyclonic system is associated with its track, speed, areal extent and saturation of catchments from prior rainfall. Rainfall totals in excess of 100 mm are common with tropical lows that move over land (BoM, 2019a).

Cyclones have affected the PU Project’s study area. The three most recent, significant cyclones affecting the Pilbara were (BoM, 2019a):

· Cyclone Bobby, 24-25 February 1995 – crossed coast just east of Onslow between midnight and 1 am on the 25th February 1995. More than 400 mm of rain fell in the Onslow area during the event. Very heavy rain associated with the cyclone caused serious flooding in the west Pilbara, Gascoyne, Goldfields and Eucla regions. Rainfall associated with this event followed heavy rains over a large part of inland WA earlier in the month.

· Cyclone Olivia, 10-11 April 1996 – crossed coast near Mardie causing wind gusts of 257 km/h before accelerating to the southeast. Pannawonica recorded gusts to 158 km/h and was extensively damaged. As Olivia passed Paraburdoo after midnight it still produced gusts to 140 km/h.

· Cyclone Monty, 1 March 2004 – passed over Mardie station west of Dampier before passing near Pannawonica where there was some damage, and the town of Pannawonica was cut-off due to flooding. Heavy rain flooded rivers. A large part of the bridge over the Maitland River on the Northwest coastal highway was washed away. 131 Air Quality Impact Assessment

Other cyclones that probably affected Burrup Peninsula weather were (sources: BoM web site): Cyclone Dominic, 22-27 January 2009; Cyclone Laurence, 16-21 December 2009; Cyclone Heidi, 9 January 2011; Cyclone Bianca, 25 January 2011; Cyclone Carlos, 14 February 2011; Cyclone Lua, 17 March 2012; Cyclone Rusty, 22 February 2013; and Cyclone Peta, 23 January 2013.

B.5 Wind Roses

Annual and seasonal wind roses created from hourly wind speed and wind direction data for BoM Karratha Aerodrome 2010-2018 are provided overleaf.

132 Air Quality Impact Assessment

133 Air Quality Impact Assessment

134 Air Quality Impact Assessment

135 Air Quality Impact Assessment

136 Air Quality Impact Assessment

137 Air Quality Impact Assessment

138 Air Quality Impact Assessment

139 Air Quality Impact Assessment

140 Air Quality Impact Assessment

141 Air Quality Impact Assessment

Appendix C. Results – Meteorological Modelling

This section provides a brief analysis of the modelling results for predicted wind speed and wind direction. The 2014 hourly datasets for the BoM weather stations at Karratha, Roebourne and Legendre Island were compared with modelled meteorological data output for the same locations, for 2014 (the simulated year used for the PU Project). The modelled predictions for wind patterns matched the observations reasonably well; annual wind roses generated from hourly data are compared in Figure C- 1. While for this 2014 case TAPM has over-predicted the frequency of south-east winds and under-predicted north-east winds, inspection of the annual and seasonal variabilities in wind direction from 2010-2018 (Appendix B) indicates the 2014 dataset is satisfactory to use as an input for an air quality impact assessment.

Figure C- 1: Annual Wind Roses Karratha 2014: TAPM (Left) and BoM Measurements (Right)

The wind speeds are compared in Table C- 1 and Figure C- 2. The comparisons show that TAPM consistently under-estimated wind speed for the Burrup Peninsula for 2014. Comparisons of results for other years indicated the problem is general, with TAPM underestimating wind speeds for other years also. While this is not ideal, nevertheless the TAPM estimates for air pollutant concentrations matched the air quality monitoring data reasonably well. Also, the use of these lower wind speeds in the modelling is considered to be a conservative step in the assessment, because the (modelled) dispersion is worse for lower wind speeds, therefore the predicted GLCs will be slightly higher.

Table C- 1: Comparisons of 2014 Hourly Average Wind Speeds

Station Karratha Aero. Roebourne Legendre Is.

TAPM TAPM TAPM Source BoM BoM BoM (1 km grid) (3 km grid) (3 km grid)

No. of averages 8755 8760 8759 8760 8756 8760

Maximum (m/s) 13.1 8.3 13.4 7.0 16.1 13.8

90th percentile (m/s) 8.0 4.6 7.8 4.3 9.7 7.2

80th percentile (m/s) 7.0 4.0 6.6 3.7 8.2 6.2

70th percentile (m/s) 6.2 3.6 5.7 3.2 7.1 5.2 142 Air Quality Impact Assessment

Station Karratha Aero. Roebourne Legendre Is.

60th percentile (m/s) 5.5 3.2 4.9 2.8 6.3 4.5

50th percentile (m/s) 4.8 2.8 4.2 2.4 5.6 3.9

Average (m/s) 4.97 2.94 4.49 2.63 5.98 4.08

Figure C- 2: Model Results for Wind Speed Compared with 2014 Observations

In the charts shown in Figure C- 2, ‘TAPM1000’ means the results were obtained from the 1000-metre resolution grid; similarly ‘TAPM3000’ refers to the 3000-metre resolution grid (Legendre Is. and Roebourne monitoring stations were outside the TAPM study area with 1 km resolution).

143 Air Quality Impact Assessment

Appendix D. Meteorological Modelling Performance

144 Memorandum

Air Quality Assessment 8th Floor, Durack Centre 263 Adelaide Terrace PO Box H615 Perth WA 6001 Australia T +61 8 9469 4400 F +61 8 9469 4488 www.jacobs.com

Subject Meteorological modelling Project Name Perdaman Urea Project performance

Attention Dan Hunter (Cardno) Project No. Jacobs: IW213400

From Matthew Pickett

Date 18 Feb 2020

Copies to Lisa Boulden, Shane Lakmaker (Jacobs)

1. Introduction

This technical memorandum sets out the results of the statistical analysis of the TAPM meteorological modelling scenario and outputs (Jacobs 2020 CTR; see References). The tasks undertaken were: § Review WA Government and Environmental Technologies & Analytics (ETA) comments on statistical analysis of TAPM meteorological output. § Review TAPM model performance based on parameters used by CSIRO scientists. § Create a list of common tests for assessment of TAPM meteorological performance. § Complete tests and provide this technical memo of results.

This Memo lists the tests used for the assessment of TAPM meteorological performance and provides the results of those tests.

It is noted the latest WA Government review comments included no review items related to performance analysis of the TAPM-generated meteorological data; see Jacobs (2020b). Other comments on the Jacobs (2019d) assessment report reviewed for this Memo are listed in the References section.

2. List of TAPM meteorological performance tests

2.1 Overview

The material set out in this section is based on a review of relevant Perdaman project communications, and the scientific literature; see References. Some background information about meteorological modelling, for the purpose of air quality impact assessment, is also provided.

2.2 Review of Purpose of Meteorological Modelling Methodology

A common purpose of an air quality assessment is to be able to predict a worst-case air quality impact in a study area; e.g., this may be the highest hourly average air pollutant concentration that would occur in any hour of a year, at any point in a study area. The result is compared with an air quality standard specific to the air pollutant. As part of this general assessment methodology, source emissions are tested on an hourly average basis using a simulated (or real) database of hourly average meteorology. This means that, in a non-leap year for example, a source is tested using 8,760 hourly

Jacobs Group (Australia) Pty Limited ABN 37 001 024 095 MP5 Memorandum

Meteorological modelling performance

meteorological simulations. After these thousands of tests, at least some of the simulated, hourly meteorological conditions are expected to match a worst-case hour in terms of air quality impact; this is why reasonable results are expected when comparing the statistical results from modelling with monitoring, as has been the case for Burrup Peninsula (Jacobs, 2019d).

While accurate modelling of hourly meteorological conditions is worthwhile; e.g., for relatively short- term air pollution forecasts, hour-by-hour accuracy is not an essential requirement for air quality impact assessment. In an air quality impact assessment, a model estimate for the worst-case air quality impact; e.g., maximum hourly average air pollutant concentration, most likely will not match the real (observed) corresponding concentration in time nor in space.

2.3 Selection of simulated meteorological year

There is no standard, rigorous, statistical procedure for determination of a representative year for modelling. Normally, statistical results for meteorological observations or modelling data; e.g., wind roses, are compared, to ensure that no year stands out as being significantly different from the norm. Primarily these annual differences are due to natural variability, but they can be due to temporary instrumental errors. Usually, comparisons of annual and seasonal wind roses, and simple comparisons of statistics between the modelled years, are more than adequate to select a simulation year for modelling.

There is often a significant amount of variability in local monthly meteorology between years; e.g., in spring, the observed winds at Karratha Aerodrome vary significantly between years (see Appendix B.5 ‘Wind Roses’ of main Jacobs report).

The original selection of the simulated meteorological year for the Woodside and Perdaman assessments was discussed in some detail during a meeting on 13 March 2019 at the Woodside Perth offices. Attendees at this meeting included ETA (Karla Hinkley), Cardno (Lorie Jones), Woodside (Robert Hearn and Anthony McMullen), and Jacobs (Michael Bell and Sarah Kelly, and Matthew Pickett by telephone). At the meeting, it was agreed the simulation year would be 2014 based on advice received by ETA and Woodside.

Selection of the simulated meteorological year, and a critical evaluation of TAPM meteorological modelling performance, were provided in the main assessment report for the Perdaman Urea Project (Jacobs, 2019d). The relevant sections were nearly identical to those provided in concurrent reports provided for the related Woodside assessment reports.

2.4 TAPM meteorological performance and list of tests

Statistical parameters commonly used by CSIRO scientists for the assessment of TAPM modelled meteorological data (hourly averages) include: mean and standard deviation, Pearson linear correlation coefficient (PCC), Root Mean Square Error (RMSE), Index of Agreement (IOA), and Skill; e.g., Physick and Blockley (2001); Hurley et al. (2004); Physick et al. (2004); Hurley et al. (2009).

The statistical tests used to assess the current, TAPM-modelled meteorological simulation for Burrup Peninsula are listed in the following points; in relation to the modelled 8,760 hourly averages of wind speed, temperature, and relative humidity, for Karratha Aerodrome. No BoM observations for Dampier were included in the analysis: The BoM Dampier Salt weather station was closed in 1993, and the BoM Dampier Port weather station closed in 2001. The statistical tests of TAPM meteorological modelling performance are described by Physick and Blockley (2001) and statistical formulae are provided by Hurley et al. (2004). The tests used for this Memo are listed in Table 2-1.

MP5 2 Memorandum

Meteorological modelling performance

Table 2-1 Key to results tables, statistics, and model performance tests

Shortening / Expansion Notes acronym

Obs. Observations BoM Karratha Aerodrome

Model N/A TAPM simulation of Burrup Peninsula meteorology; 8,760 hourly averages; grid resolution 1000m

N Number, or population Number of hourly averages used for comparisons

Mean N/A Arithmetic mean of hourly averages

SD Standard Deviation SD of hourly averages

PCC Pearson or linear 0=no correlation; 1=perfect correlation Correlation Coefficient

RMSE Root Mean Square Error Standard deviation of prediction errors around line of best fit

IOA Index of Agreement Wilmott IOA: 0=no agreement, 1=perfect agreement

SkillV N/A Ratio of modelled to observed standard deviation; should be close to unity

SkillR N/A Ratio of RMSE to observed standard deviation; should be less than unity

3. Results

The results of the statistical tests of TAPM meteorological modelling performance are set out in the following tables. Basic statistics are set out for each of the parameters wind speed, temperature and relative humidity. The results are continued in the second table in each case, containing PCC, IOA, etc.

Table 3-1 TAPM results for Karratha; basic statistics – wind speed (m/s)

Dataset N Mean Obs. Mean Model SD Obs. SD Model

CSIRO, 1999 met. 8737 4.4 4.5 2.3 1.8

Jacobs, 2012 met. 8782 5.1 2.9 2.3 1.3

Jacobs, 2014 met. 8757 5.0 2.9 2.3 1.3

Jacobs, 2018 met. 8751 5.4 3.1 2.5 1.3

Table 3-2 TAPM results for Karratha; model performance tests – wind speed

Dataset PCC RMSE IOA SkillV SkillR

CSIRO, 1999 met. 0.65 1.79 0.79 0.79 0.77

Jacobs, 2012 met. -0.01* 3.44 0.41 0.54 1.47

Jacobs, 2014 met. 0.65 2.71 0.61 0.54 1.16

Jacobs, 2018 met. 0.69 2.89 0.61 0.54 1.17 *Correlation is poor between wind speed data for 2012; however, the basic statistics are similar to 2014.

Table 3-3 TAPM results, results for Karratha; basic statistics – temperature (oC)

Dataset N Mean Obs. Mean Model SD Obs. SD Model

CSIRO, 1999 met. 8737 25.6 25.0 5.0 6.3

Jacobs, 2012 met. 8763 26.0 26.0 6.0 4.7

Jacobs, 2014 met. 8757 26.3 26.3 6.2 4.9

MP5 3 Memorandum

Meteorological modelling performance

Dataset N Mean Obs. Mean Model SD Obs. SD Model

Jacobs, 2018 met. 8751 25.9 25.9 5.8 5.0

Table 3-4 TAPM results for Karratha; model performance tests – temperature

Dataset PCC RMSE IOA SkillV SkillR

CSIRO, 1999 met. 0.92 2.69 0.94 1.25 0.54

Jacobs, 2012 met. 0.92 3.10 0.92 0.78 0.81

Jacobs, 2014 met. 0.94 3.10 0.92 0.79 0.50

Jacobs, 2018 met. 0.94 2.07 0.96 0.85 0.35

Table 3-5 TAPM results for Karratha; basic statistics – relative humidity (%)

Dataset N Mean Obs. Mean Model SD Obs. SD Model

CSIRO, 1999 met. 8736 60.3 56.0 21.9 21.3

Jacobs, 2012 met. 8748 58.5 53.9 20.6 18.7

Jacobs, 2014 met. 8756 59.2 52.8 21.4 18.1

Jacobs, 2018 met. 8749 55.8 50.9 19.4 18.6

Table 3-6 TAPM results for Karratha; model performance tests – relative humidity

Dataset PCC RMSE IOA SkillV SkillR

CSIRO, 1999 met. 0.79 14.56 0.88 0.97 0.66

Jacobs, 2012 met. 0.71 16.60 0.82 0.91 0.81

Jacobs, 2014 met. 0.66 18.26 0.78 0.84 0.85

Jacobs, 2018 met. 0.72 15.28 0.83 0.96 0.79

4. Discussion of wind speed performance

The results of tests for TAPM wind speed performance, (Table 3-1 and Table 3-2 ), show that the current TAPM modelling of wind speed, for all the current simulations, is poorer than for the 1999 simulation investigated by CSIRO scientists in the 2000s; e.g., Hurley et al. (2004). The modification in TAPM V.4 that seems likely to have made the largest contribution to these discrepancies was new boundary layer physics for low wind speeds during very stable atmospheric conditions; e.g., see Hurley et al. (2009). However, Hurley et al. (2009) reported the TAPM V.4 changes improved predictions of near-surface wind speeds for the complex situation of low wind speeds at night. TAPM V.4 performance tests were reported by Hurley et al. (2009) for six locations, including for Perth and Kwinana (results at 10-metre heights). These indicated good results were obtained for wind speed (PCC: 0.65-0.86; IOA: 0.77-0.92), indicating that the current discrepancies between wind speed for the Burrup Peninsula may not be a fault of the model version.

Other differences between the previous CSIRO and Jacobs (2016b) model setups that may have contributed to the discrepancies in wind speed were: the 1999 simulation by CSIRO used a BoM Global Analysis and Prediction (GASP) meteorological database as input, whereas Jacobs (2016b) used a U.S. National Centers for Environmental Prediction (NCEP) dataset; i.e., the standard database for use with TAPM V.4 supplied by the CSIRO. Also, there were some different selections for land type, which affects surface roughness settings, for example. It seems likely further improvements could be made

MP5 4 Memorandum

Meteorological modelling performance

to TAPM V.4 meteorological modelling for the Burrup Peninsula, by further experimentation with the model.

5. Summary

As is already known and reported, the TAPM simulations underestimated wind speed for the current 2014 simulation case; e.g., see Jacobs (2019b). The correlation between the wind speed data for the 2012 simulation was particularly poor, due to particularly bad correlation from February to May in 2012. However, the basic statistics of the 2012 simulation were similar to those for the 2014 simulation. The other tests indicate the current TAPM predictions for wind speed on Burrup Peninsula did not perform as well as the 1999 meteorological simulation generated by the CSIRO; e.g., Hurley et al. (2004), and others (see References).

The current TAPM results for temperature are very good (2012, 2014 and 2018), and on a par with the CSIRO simulations of 1999 meteorology for Burrup Peninsula. The current results for relative humidity are satisfactory, although TAPM did not perform as well for the 2014 simulation.

The TAPM 2014 and 2018 meteorological simulations for Burrup Peninsula are considered satisfactory for the purpose of air quality impact assessment. The poor correlation of wind speeds in the 2012 simulation means further review of that simulation would be advisable, prior to its use in an assessment.

As a final note, the reasons for the decision to use the latest version of TAPM (V.4) for the Woodside and Perdaman assessments was discussed in Section 5.2 of Jacobs (2019b). There are sound, scientific reasons for using the latest version of the model. TAPM V.4 included a substantial number of new improvements to the modelled physics; e.g., see Hurley et al. (2009). Also, corrections and other changes may have been made to TAPM V.4 that were undocumented. While a previous version of the model may have delivered more accurate results in some circumstances, it would have been for the wrong reasons.

References:

Jacobs (2019a): Jacobs Memorandum, Information for Peer Review, Perdaman Urea Project, From Matthew Pickett, Attention Maelle Bourdais, Copies to Lisa Boulden, Daniel Hunter, Jon Harper, Project No. IW213400, 1 August 2019.

Jacobs (2019b): Jacobs Memorandum, Jacobs Response to Initial Peer Review Comments, Perdaman Urea Project, From Lisa Boulden, Attention Maelle Bourdais, Copies to Matt Pickett, Daniel Hunter, Jon Harper, Project No. IW213400, 13 August 2019.

Jacobs (2019c): Jacobs MS Spreadsheet, PUP AQ Report-Revision 1-Comments Tracking Sheet_20191001_ETA.xlsx, containing Jacobs responses to ETA’s second review, 1 October 2019.

Jacobs (2019d): Jacobs, Perdaman Urea Project, Cardno (WA) Pty Ltd, Air Quality Impact Assessment, Final | Revision 4, 13 November 2019.

Jacobs (2020a): Jacobs File Note: DMA comment re: use of TAPM-GRS, 22 January 2020.

Jacobs (2020b): Jacobs MS Spreadsheet, ERD_Review_Comments_collated_Jacobs_MP4_LB1.xlsx, containing Jacobs responses to ERD’s comments “v.1.2”, 23 January 2020.

MP5 5 Memorandum

Meteorological modelling performance

Jacobs (2020c): Jacobs Cost, Time, Resource (CTR) 19-01-V3, Perdaman Urea Plant – Air Quality Modelling – Variation for Responding to EPA Comments, 4 February 2020.

P.J. Hurley, W.L. Physick and M.E. Cope, Summary of TAPM Verification for the Pilbara Region, CSIRO Atmospheric Research, A Report to the Department of Environment, WA, March, 2004.

P. Hurley, M. Edwards and A. Luhar, Evaluation of TAPM V4 for several meteorological and air pollution datasets, Air Quality and Climate Change, Vol.43, No.3, pp.19-24, August 2009.

W. Physick and A. Blockley, An Evaluation of Air Quality Models for the Pilbara Region, CSIRO Atmospheric Research, June 2001.

W.L. Physick, K.N. Rayner, P. Mountford, and M. Edwards, Observations and modelling of dispersion meteorology in the Pilbara region, Aust. Met. Mag. 53 (2004), pp.175-187, CSIRO Atmospheric Research, Aspendale, Australia, Manuscript received July 2003; revised May 2004.

MP5 6 Air Quality Impact Assessment

Appendix E. Discrete Receptor Results – Air Quality

Table E-1: Summary of Results PNO: Discrete Receptor Maxima and Standards

Assessment Parameter & AQMS Grid Max. Grid Max. Air Quality AQMS AQMS Hearson Standing Grid Burrup Ngajarli MNP-CN MNP-SE King Bay MLKC location location Standard Dampier Karratha Cove Stones Maximum Road E (m) N (m) [standard µg/m3]

Max 1h NO2

(NO2/NOx 100%); 18.5 5.3 5.9 15.0 14.2 8.4 12.5 16.9 13.0 11.5 31.7 476,489 7,718,717 [226]

Max 1h NO2

(NO2/NOx 30%) 5.5 1.6 1.8 4.5 4.3 2.5 3.8 5.1 3.9 3.5 9.5 476,489 7,718,717 [226]

Annual NO2 [56] 0.51 0.11 0.11 1.08 1.25 0.19 0.21 0.81 0.35 0.22 3.3 476,489 7,718,717

Annual NO2

(NO2/NOx 30%) 0.15 0.03 0.03 0.32 0.37 0.06 0.06 0.24 0.10 0.07 1.0 476,489 7,718,717 [56]

Annual NO2/NOx 30% and capacity 0.14 0.03 0.03 0.29 0.34 0.05 0.06 0.22 0.09 0.06 0.9 476,489 7,718,717 factor 90.6% [56]

Max 1h SO2 200 0.3 0.1 0.1 0.2 0.2 0.1 0.2 0.3 0.2 0.2 0.4 476,489 7,718,717 ppb at 25 °C [524]

Max 24h SO2 80ppb at 25 °C 0.07 0.02 0.01 0.07 0.09 0.03 0.03 0.07 0.03 0.03 0.12 476,489 7,718,717 [209]

145 Air Quality Impact Assessment

Assessment Parameter & AQMS Grid Max. Grid Max. Air Quality AQMS AQMS Hearson Standing Grid Burrup Ngajarli MNP-CN MNP-SE King Bay MLKC location location Standard Dampier Karratha Cove Stones Maximum Road E (m) N (m) [standard µg/m3]

Annual SO2 20ppb 0.008 0.002 0.002 0.017 0.020 0.003 0.003 0.013 0.005 0.004 0.05 476,489 7,718,717 at 25 °C [52.4]

Annual SO2 with capacity factor 0.008 0.002 0.001 0.015 0.018 0.003 0.003 0.012 0.005 0.003 0.05 476,489 7,718,717 90.6% [52.4]

Max 24h PM10 [50] 5.6 1.4 1.6 6.7 7.5 1.7 3.3 7.2 4.3 3.1 12.4 476,489 7,718,717

Annual PM10 [25] 0.80 0.14 0.13 1.82 2.06 0.22 0.28 1.41 0.46 0.30 7.0 476,489 7,718,717

Annual PM10 with capacity 0.72 0.12 0.12 1.65 1.86 0.20 0.25 1.28 0.42 0.27 6.4 476,489 7,718,717 factor (90.6%) [25]

Max 24h PM2.5 (PM10 ratio = 1.4 0.4 0.4 1.7 1.9 0.4 0.8 1.8 1.1 0.8 3.7 476,489 7,718,717 25%) [25]

Annual PM2.5 including CF and 0.2 0.0 0.0 0.4 0.5 0.1 0.1 0.3 0.1 0.1 1.9 476,489 7,718,717 PM ratio [8]

Max 1h Ammonia 33.7 17.0 8.7 33.5 35.0 9.9 31.5 34.2 36.6 24.6 76.9 476,489 7,718,717 (NH3) [330]

Max 1h Methanol 0.01 0.00 0.00 0.01 0.01 0.01 0.01 0.02 0.01 N/A 0.02 476,489 7,718,717 [3000]

Max 1h CH2O [20] 0.11 0.03 0.03 0.08 0.08 0.05 0.07 0.11 0.10 N/A 0.14 476,489 7,718,717

146 Air Quality Impact Assessment

Assessment Parameter & AQMS Grid Max. Grid Max. Air Quality AQMS AQMS Hearson Standing Grid Burrup Ngajarli MNP-CN MNP-SE King Bay MLKC location location Standard Dampier Karratha Cove Stones Maximum Road E (m) N (m) [standard µg/m3]

Max 24h CH2O 0.02 0.01 0.01 0.03 0.03 0.01 0.01 0.03 0.01 N/A 0.04 476,489 7,718,717 [49]

Table E-2: Summary of Results PUC: Discrete Receptor Maxima and Standards

Assessment Parameter & AQMS Grid Max. Grid Max. Air Quality AQMS AQMS Hearson Standing Grid Burrup Ngajarli MNP-CN MNP-SE King Bay MLKC location E location N Standard Dampier Karratha Cove Stones Maximum Road (m) (m) [standard µg/m3]

Max 1h NO2

(NO2/NOx 100%) 15.0 4.0 4.5 10.8 10.5 6.7 8.8 12.2 9.1 7.8 24.2 476,489 7,718,717 [226]

Max 1h NO2

(NO2/NOx 30%) 4.5 1.2 1.4 3.3 3.2 2.0 2.6 3.7 2.7 2.4 7.2 476,489 7,718,717 [226]

Max 1h SO2 200 0.2 0.1 0.1 0.2 0.2 0.1 0.1 0.2 0.1 0.1 0.37 476,489 7,718,717 ppb at 25 °C [524]

Max 24h SO2 80ppb at 25 °C 0.05 0.01 0.01 0.06 0.07 0.02 0.02 0.06 0.02 0.03 0.11 476,489 7,718,717 [209]

147 Air Quality Impact Assessment

Assessment Parameter & AQMS Grid Max. Grid Max. Air Quality AQMS AQMS Hearson Standing Grid Burrup Ngajarli MNP-CN MNP-SE King Bay MLKC location E location N Standard Dampier Karratha Cove Stones Maximum Road (m) (m) [standard µg/m3]

Max 24h PM10 [50] 8.2 1.8 2.1 9.4 10.3 2.1 4.0 10.7 5.4 4.2 19.1 476,489 7,718,717

Max 24h PM2.5

(PM10 ratio = 25%) 2.1 0.4 0.5 2.3 2.6 0.5 1.0 2.7 1.4 1.0 5.7 476,489 7,718,717 [25]

Max 1h Ammonia 31.5 16.2 8.7 28.2 27.8 11.5 22.7 32.6 34.4 26.0 75.9 476,489 7,718,717 (NH3) [330]

Table E-3: Summary of Results Baseline: Discrete Receptor Maxima and Standards

Air Quality AQMS Grid Max. Grid Max. Standard AQMS AQMS Hearson Standing Grid Burrup Ngajarli MNP-CN MNP-SE King Bay location E location N (units) Dampier Karratha Cove Stones Maximum Road (m) (m) [standard]

Max 1h NO2 (ppb) [120] 34.5 24.8 24.9 36.6 33.4 24.4 30.0 33.6 30.5 42.6 476,489 7,722,717

Annual without CF 3.2 1.7 0.9 3.1 3.6 1.3 1.3 2.6 2.4 5.0 477,489 7,718,717 NO2 (ppb) [30]

Max 1h O3 (ppb) [100] 58.8 55.4 58.2 55.0 56.1 59.0 57.4 59.2 60.3 61.8 472,489 7,718,717

Max 4h O3 (ppb) [80] 54.3 52.6 56.6 49.7 51.4 56.1 51.6 56.0 56.7 58.2 472,489 7,718,717

148 Air Quality Impact Assessment

Air Quality AQMS Grid Max. Grid Max. Standard AQMS AQMS Hearson Standing Grid Burrup Ngajarli MNP-CN MNP-SE King Bay location E location N (units) Dampier Karratha Cove Stones Maximum Road (m) (m) [standard]

Max 1h SO2 (ppb) [200] 11.2 13.2 3.6 9.2 9.5 7.3 8.7 9.3 10.9 18.2 470,489 7,717,717

Max 24h SO2 (ppb) [80] 4.7 4.5 1.7 4.0 3.5 2.3 3.0 4.2 5.0 7.0 470,489 7,717,717

Annual SO2 without CF (ppb) 2.0 1.6 0.9 1.4 1.3 0.9 1.1 1.7 1.9 4.5 470,489 7,717,717 [20]

Max 24h PM10 (µg/m3) [25] 34.4 34.5 34.1 34.4 34.3 33.9 34.2 34.5 34.4 35.5 471,489 7,703,717

Annual PM10 3 without CF (µg/m ) 23.8 23.7 23.8 23.8 23.8 23.3 23.7 23.9 23.9 24.8 470,489 7,717,717 [25]

Max 24h PM2.5 (µg/m3) [8] 15.0 15.3 14.5 14.9 15.0 14.5 14.6 15.0 14.9 15.5 470,489 7,717,717

Annual PM2.5 3 without CF (µg/m ) 8.1 7.9 7.9 8.0 8.0 7.9 7.9 8.0 8.0 8.4 470,489 7,717,717 [8]

Max 1h Ammonia 0.8 0.7 0.9 1.1 0.9 1.0 1.1 0.1 1.2 0.4 476,489 7,718,717 (NH3) [330]

149 Air Quality Impact Assessment

Table E-4: Summary of Results BPUC: Discrete Receptor Maxima and Standards

Air Quality Standard AQMS Grid Max. Grid Max. AQMS AQMS Hearson Standing MLKC Max. on Burrup Ngajarli MNP-CN MNP-SE King Bay location E location N (units) Dampier Karratha Cove Stones Grid Road (m) (m) [standard]

Max 1h NO2 (ppb) [120] 33.5 24.8 25.4 36.9 33.7 25.4 31.2 34.2 31.1 19.8 42.9 476,489 7,722,717

Max 1h O3 (ppb) [100] 58.7 55.4 58.4 55.1 56.1 59.2 57.3 58.1 60.3 59.2 61.9 472,489 7,718,717

Max 4h O3 (ppb) [80] 54.1 52.6 56.7 49.2 51.4 56.2 51.6 54.7 56.8 56.5 58.2 472,489 7,718,717

Max 1h SO2 (ppb) [200] 11.3 12.9 3.6 9.2 9.6 7.4 8.4 10.5 10.9 10.0 18.1 470,489 7,717,717

Max 24h SO2 (ppb) [80] 4.7 4.6 1.7 4.0 3.5 2.3 3.0 4.1 5.0 2.9 7.0 470,489 7,717,717

Max 24h PM10 (µg/m3) [50] 37.9 34.7 34.5 41.7 42.4 34.5 36.3 39.6 36.1 35.2 53.0 476,489 7,718,717

Max 24h PM2.5 (µg/m3) [25] 15.2 15.5 14.8 16.6 16.5 14.7 15.0 15.9 15.5 14.7 18.9 476,489 7,718,717

Max 1h Ammonia 3 (NH3) (µg/m ) 32.0 16.6 9.1 28.9 28.0 12.3 22.9 34.5 35.3 27.1 76.2 476,489 7,718,717 [330]

150 Air Quality Impact Assessment

Table E-5: Summary of Results BPNO: Discrete Receptor Maxima and Standards

Air Quality AQMS Grid Max. Grid Max. Standard AQMS AQMS Hearson Standing Max. on Burrup Ngajarli MNP-CN MNP-SE King Bay MLKC location E location N (units) Dampier Karratha Cove Stones Grid Road (m) (m) [standard]

Max 1h NO2 (ppb) [120] 33.6 24.8 25.6 37.0 33.7 25.7 31.6 34.1 31.5 20.6 43.1 476,489 7,722,717

Annual NO2 (ppb) [30] 3.3 1.7 0.9 3.4 4.0 1.4 1.3 3.0 2.5 1.7 5.6 477,489 7,718,717

Max 1h O3 (ppb) [100] 58.7 55.4 58.6 55.3 56.3 59.1 57.3 58.0 60.4 59.2 62.0 472,489 7,718,717

Max 4h O3 (ppb) [80] 54.0 52.7 56.9 49.1 51.5 56.2 51.6 54.7 56.8 56.5 58.3 472,489 7,718,717

Max 1h SO2 (ppb) [200] 11.3 12.9 3.6 9.2 9.6 7.4 8.4 10.5 10.9 10.0 18.1 470,489 7,717,717

Max 24h SO2 (ppb) [80] 4.7 4.6 1.7 4.0 3.5 2.3 3.0 4.1 5.0 2.9 7.0 470,489 7,717,717

Annual SO2 (ppb) [20] 2.0 1.6 0.9 1.4 1.3 0.9 1.1 1.7 1.9 1.1 4.5 470,489 7,717,717

Max 24h PM10 (µg/m3) [50] 36.6 34.6 34.4 39.2 39.6 34.2 35.4 37.6 35.5 34.6 44.7 476,489 7,718,717

Annual PM10 (µg/m3) [25] 24.5 23.8 23.9 25.5 25.8 23.5 24.0 25.1 24.4 23.8 30.9 476,489 7,718,717

Max 24h PM2.5 (µg/m3) [25] 15.2 15.5 14.7 16.0 15.9 14.7 14.9 15.6 15.4 14.7 17.4 476,489 7,718,717

151 Air Quality Impact Assessment

Air Quality AQMS Grid Max. Grid Max. Standard AQMS AQMS Hearson Standing Max. on Burrup Ngajarli MNP-CN MNP-SE King Bay MLKC location E location N (units) Dampier Karratha Cove Stones Grid Road (m) (m) [standard]

Annual PM2.5 (µg/m3) [8] 8.3 8.0 7.9 8.6 8.7 7.9 8.0 8.4 8.2 8.0 10.3 476,489 7,7187,17

Max 1h Ammonia (NH3) (µg/m3) 34.2 17.4 9.1 34.2 35.2 10.7 31.8 36.2 37.4 25.7 77.3 476,489 7,718,717 [330]

Table E-6: Summary of Results FPNO: Discrete Receptor Maxima and Standards

Air Quality AQMS Grid Max. Grid Max. Standard AQMS AQMS Hearson Standing Max. on Burrup Ngajarli MNP-CN MNP-SE King Bay MLKC location E location N (units) Dampier Karratha Cove Stones Grid Road (m) (m) [standard]

Max 1h NO2 (ppb) [120] 34.2 25.8 28.4 37.7 35.4 30.2 32.9 36.0 33.9 25.5 43.9 476,489 7,722,717

Annual NO2 (ppb) [30] 4.0 1.8 1.0 3.7 4.4 1.6 1.4 3.7 2.7 1.9 5.9 477,489 7,718,717

Max 1h O3 (ppb) [100] 58.4 56.5 61.2 56.1 57.7 59.3 57.8 58.1 61.3 58.7 63.0 472,489 7,718,717

Max 4h O3 (ppb) [80] 53.7 53.6 59.1 49.1 51.7 57.0 52.2 54.7 57.8 56.7 59.7 492,489 7,707,717

Max 1h SO2 (ppb) [200] 11.4 12.9 3.6 9.2 9.6 7.4 8.4 10.6 10.9 10.0 18.1 470,489 7,717,717

152 Air Quality Impact Assessment

Air Quality AQMS Grid Max. Grid Max. Standard AQMS AQMS Hearson Standing Max. on Burrup Ngajarli MNP-CN MNP-SE King Bay MLKC location E location N (units) Dampier Karratha Cove Stones Grid Road (m) (m) [standard]

Max 24h SO2 (ppb) [80] 4.8 4.6 1.7 4.0 3.5 2.3 3.0 4.1 5.0 2.9 7.0 470,489 7,717,717

Annual SO2 (ppb) [20] 2.0 1.6 0.9 1.4 1.3 0.9 1.1 1.7 1.9 1.1 4.5 470,489 7,717,717

Max 24h PM10 (µg/m3) [50] 36.6 34.7 34.4 39.3 39.6 34.2 35.5 37.6 35.6 34.6 44.6 476,489 7,718,717

Annual PM10 (µg/m3) [25] 24.6 23.8 23.9 25.6 25.8 23.5 24.0 25.1 24.4 23.8 30.8 476,489 7,718,717

Max 24h PM2.5 (µg/m3) [25] 15.3 15.5 14.8 16.1 16.0 14.7 15.0 15.8 15.5 14.7 17.4 476,489 7,718,717

Annual PM2.5 (µg/m3) [8] 8.3 8.0 7.9 8.6 8.7 7.9 8.0 8.5 8.2 8.0 10.3 476,489 7,718,717

Max 1h Ammonia (NH3) (µg/m3) 34.2 17.4 9.1 34.2 35.2 10.7 31.8 36.2 37.4 25.7 77.3 476,489 7,718,717 [330]

153 Air Quality Impact Assessment

Appendix F. Discrete Receptor Results – Vegetation

Table F-1: Summary of Results PNO: Discrete Receptor Maxima and EU 2008 Standards

Assessment Parameter AQMS AQMS AQMS Hearson Standing Grid Ngajarli MNP-CN MNP-SE King Bay MLKC [EU 2008 Burrup Road Dampier Karratha Cove Stones Maximum Standard]

Annual NOx [30 3.30 µg/m3] 0.51 0.11 0.11 1.08 1.25 0.19 0.21 0.81 0.35 0.22

Annual NOx with capacity 0.47 0.10 0.10 0.98 1.13 0.17 0.19 0.74 0.31 0.20 2.99 factor (90.6%) [30 µg/m3]

Annual SO2 [20 0.008 0.002 0.002 0.017 0.020 0.003 0.003 0.013 0.005 0.004 0.053 µg/m3]

Annual SO2 with capacity 0.008 0.002 0.001 0.015 0.018 0.003 0.003 0.012 0.005 0.003 0.048 factor (90.6%) [30 µg/m3]

154 Air Quality Impact Assessment

Table F-2: Summary of Results Baseline: Discrete Receptor Maxima and EU 2008 Standards

Assessment Parameter & AQMS Burrup AQMS AQMS Standing Grid Ngajarli Hearson Cove MNP-CN MNP-SE King Bay EU 2008 Road Dampier Karratha Stones Maximum Standard

Annual NOx (30 µg/m3, T = 4.0 2.1 1.1 4.2 4.4 1.6 1.5 3.3 2.9 7.7 30oC); 16.2 ppb

Annual SO2 (20 µg/m3, T = 2.0 1.6 0.9 1.4 1.3 0.9 1.1 1.7 1.9 4.5 30oC); 7.8 ppb

Table F-3: Summary of Results BPNO: Discrete Receptor Maxima and EU 2008 Standards

Assessment Parameter AQMS AQMS AQMS Hearson Standing Grid Ngajarli MNP-CN MNP-SE King Bay MLKC & EU 2008 Burrup Road Dampier Karratha Cove Stones Maximum Standard

Annual NOx (30 µg/m3, T = 4.2 2.1 1.1 4.6 5.0 1.7 1.6 4.0 3.0 2.1 8.6 30oC); 16.2 ppb

Annual SO2 (20 µg/m3, T = 2.0 1.6 0.9 1.4 1.3 0.9 1.1 1.7 1.9 1.1 4.5 30oC); 7.8 ppb

155 Air Quality Impact Assessment

Table F-4: Summary of Results FPNO: Discrete Receptor Maxima and EU 2008 Standards

Assessment Parameter AQMS AQMS AQMS Hearson Standing Grid Ngajarli MNP-CN MNP-SE King Bay MLKC & EU 2008 Burrup Road Dampier Karratha Cove Stones Maximum Standard

Annual NOx (30 µg/m3, T = 5.2 2.3 1.3 5.1 5.7 2.0 1.8 5.2 3.3 2.3 9.0 30oC); 16.2 ppb

Annual SO2 (20 µg/m3, T = 2.0 1.6 0.9 1.4 1.3 0.9 1.1 1.7 1.9 1.1 4.5 30oC); 7.8 ppb

156