NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to 2040

Transcription

1 NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to 2040

2 Executive Summary To ensure that tunnel ventilation systems are accurately designed, it is critical that key design inputs are appropriately defined for each project. The vehicle emission factors that are utilised in the sizing of ventilation systems in road tunnels are one such input. The World Road Association (PIARC) released a report in 2012 titled Road Tunnels: Vehicle Emissions and Air Demand for Ventilation. The report provides data for calculating vehicle emission factors, namely for the exhaust pollutants CO, NO x and Particulate Matter. These factors have been primarily calculated for the European vehicle Fleet and are currently approximated for a generalised Australian vehicle Fleet up to This report provides a revised, ground-up approach to determining the vehicle emission factors specifically for the NSW vehicle Fleet. This approach is acknowledged by PIARC (2012) and is termed the detailed method. By analysing the composition of the NSW vehicle Fleet between 2000 and 2015, aligning the data with the emission standards and weighting the Fleet for vehicle kilometres travelled, emission factors have been forecasted from 2016 to It is proposed that the methodology and emissions factors presented here will be applied in the design of tunnel ventilation systems for future road tunnel projects. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to 2040 II

3 Contents 1 Introduction Objectives of this report Background Simplified Method for Estimating Australia s Vehicle Emissions Overview Limitations NSW Fleet Analysis Overview Stage 1: NSW Registration Fleet Forecast Part A: Information Gathering Discussion on Fuel and Vehicle Type Classifications Part B: Registration Fleet Forecasting Methodology Part C: Alignment with Vehicle Emission Standards Conclusion Stage 2: NSW VKT-weighted Fleet Forecast Part A: Information Gathering Part B: VKT Weighted Fleet Forecasting Methodology NSW Fleet Forecast for Ventilation Design References Appendix A Case Study: NSW Emission Factors A.1 Overview A.2 Methodology A.3 Primary NO 2 :NO x Analysis DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to 2040 III

4 Author: Reviewed by: Daniel O Kelly, Scholar, Tunnel Technology, Network Motorways Nigel Casey, Manager, Tunnel Technology, Network Motorways Adam Garland, Senior Engineer, Tunnel Technology, Network Motorways Date: September 2016 Version: DRAFT Reference: Division: Roads and Maritime Services Transport for NSW DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to 2040 IV

5 Terms and Definitions Term ADR Age Distribution Base emission factor Class CO Description Australian Design Rules The spread of vehicle emission standards associated with vehicles of a particular class, based on the year in which the specific vehicle was manufactured A quantity which expresses the mass or amount of a pollutant emitted per unit of time (e.g. for CO emissions, the base emission factors are measured in grams per hour (g/h)) A specific vehicle and fuel type combination (e.g. Petrol Passenger ) Carbon Monoxide Euro Standard (Euro) The European vehicle emission standards (e.g. Euro 1) Fleet Future Year Influencing Factor GMR HBEFA HGV LDV NO2 NOx NSW EPA Particulate Matter (PM) PC Percentage Mix PHEM PIARC PIARC Detailed Method PIARC Simple Method Pollutants RMS Vehicle Emission Standard VKT The breakdown of all vehicles in NSW based on different vehicle classes (Passenger, Light Duty, and Heavy Goods ), Fuel Types (Petrol, Diesel) and emissions standards (Euro 1, Euro 2, etc.) A fraction that adjusts the emission factors depending on the ventilation design year Greater Metropolitan Region Handbook of Emission Factors for Road Transport Heavy Goods Vehicle Light-Duty Vehicle Nitrogen Dioxide Oxides of Nitrogen New South Wales Environment Protection Authority Suspended particles in the air Passenger Cars The distribution between two categories, expressed as a percentage (e.g. between petrol and diesel fuelled vehicles) Passenger Car and Heavy Duty Emission Model World Road Association Application of a specific vehicle Fleet to corresponding Euro emission factors, as provided in the PIARC documentation The Australian Fleet emission factors, as provided in the PIARC documentation Refers to the emissions of CO, NOx, NO2 and PM for this report Roads and Maritime Services Emission laws that outline the maximum levels of pollutants Vehicle Kilometres Travelled DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to 2040 V

6 1 Introduction 1.1 Objectives of this report The intent of this report is to provide a validated forecast of the on road, NSW vehicle Fleet, taking into account trends in vehicle registrations, vehicle age and vehicle kilometers travelled. The NSW Fleet forecast will enable tunnel ventilation designers to accurately estimate vehicle emissions for future road tunnel projects up to the year 2040, by supplementing the PIARC (2012) methodology. This report outlines the methodology and assumptions in trending the existing Fleet for the years 2000 to 2015 and the forecasted Fleet up until Appendix A provides an example application of the NSW Fleet Forecast to estimate NSW-specific emission factors using the PIARC (2012) Methodology. Analysis is also conducted into the levels of primary NO2 as a percentage of NOx. 1.2 Background To ensure that tunnel ventilation system capacities are accurately calculated and input parameters are robustly substantiated, it is critical that key design inputs are appropriately defined for each project. The vehicle emission factors that are utilised in the sizing of ventilation systems in road tunnels are one such input. In 2012, the World Road Association (PIARC) released a report titled Road Tunnels: Vehicle Emissions and Air Demand for Ventilation. This report defines the minimum air demand that is required to adequately ensure in-tunnel air quality and visibility thresholds are met. The three main exhaust pollutants that PIARC (2012) provide emission factors for are CO, NO x and particulate matter. When calculating the air demand in road tunnels, these emission factors determine the volume of fresh air required to adhere to health and safety, environmental and visibility guidelines. Consequently, this impacts the sizing of tunnel ventilation equipment. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

7 1.3 Simplified Method for Estimating Australia s Vehicle Emissions Overview Australian road tunnels designed or delivered in recent years have used the PIARC simple method for determining vehicle emissions factors. PIARC (2012) sets out this methodology through the use of Equation 1. Equation 1: Q = (q ex (v,i) fh ft fe fm (HGV only) ) + q ne (v) Q q ex (v, i) emission for CO, NOx [g/(h.veh)] and emissions of particle matter [m 2 /(h.veh)] base emission factor for PC/LDV/HGV with gasoline or diesel engines, depending on average speed and road gradient for the base year 2010 [g/(h.veh)] or [m 2 /(h.veh)] q ne (v) emission factor for non-exhaust particulate emissions [m 2 /(h.veh)] fh altitude factor [-] ft future year influencing factor from the base year [ -] fe influencing factor for other technology standards [ -] fm influencing factor for vehicle gross masses [ - ] (HGVs only) The base emission factor (denoted in Equation 1 as q ex (v,i) ) is defined for a certain average speed and road gradient in the base year These values are presented in PIARC (2012) as shown in Table 1, for speeds from 0 km/h to 130 km/h and gradients from -6% to +6%. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

8 Table 1: CO emissions (g/h) from the Australian Fleet of petrol passenger vehicles for Information obtained from PIARC (2012) Table 29, p39 Gradient (%) v [km/h] , , , , , , ,163.5 Emission factors for the Australian fleet are provided in PIARC (2012) for CO, NOx and particulate matter (PM) for petrol passenger vehicles, diesel passenger vehicles, a petrol/diesel mix of light-duty vehicles and diesel heavy goods vehicles. The information provided in Table 1, and subsequent tables for different pollutants and vehicle classes, are for the base year PIARC (2012) provides a future year influencing factor to account for more stringent vehicle emission standards that have been implemented since The PIARC simple method is to apply the relevant future year influencing factor (as presented in Table 2) to the base emission factor in the required design year. The future year influencing factor is denoted as (ft) in Equation 1. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

9 Table 2: Future Year Influencing Factor (ft) for years different to the base-year for Australia Passenger. Information obtained from PIARC (2012) Table 34, p42 Passenger CO NO x Opacity ft Petrol Diesel Petrol Diesel Diesel Not provided Not provided Not provided Not provided Not provided 2030 Not provided Not provided Not provided Not provided Not provided The altitude factor (fh) and the influencing factor for different technology standards (fe) are not considered to be applicable for Australia. The consolidated equation for estimating vehicle emission factors in Australia for future projects is therefore shown in Equation 2. According to PIARC (2012), the emission factors for non-exhaust particulate emissions (q ne (v)) are proportional to the number of vehicles in the tunnel [and] are only dependent on the type of vehicle, not on the model year. Equation 2: Q = (q ex (v,i) ft fm (HGV only) ) + q ne (v) Q q ex (v, i) emission for CO, NOx [g/(h.veh)] and emissions of particle matter [m 2 /(h.veh)] base emission factor for PC/LDV/HGV with gasoline or diesel engines, depending on average speed and road gradient for the base year 2010 [g/(h.veh)] or [m 2 /(h.veh)] q ne (v) emission factor for non-exhaust particulate emissions [m 2 /(h.veh)] ft future year influencing factor from the base year [ -] fm influencing factor for vehicle gross masses [ - ] (HGVs only) DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

10 1.3.2 Limitations The PIARC (2012) simple method, used to determine Australia s vehicle emission factors, is not completely transparent in the information provided within PIARC (2012). PIARC (2012) states that questionnaires concerning Fleet distribution data and emission standards (implementation years) were sent to national delegates of PIARC and other institutions to collect country-specific vehicle-emission information and based on this, the base emission factors were determined. The simplified method that has been imposed on the Australian Fleet is calculated depending on the percentage of Pre-Euro (before the first emission standard) vehicles that exist in the reference year Table 3 presents this data. Table 3: Criteria for Technology Standard A, B and C category selection. Information obtained from PIARC (2012) Table 25, p34 Reference Year: 2010 Percentage of Pre-Euro 1 Type Standard A Standard B Standard C PC Petrol < 20% 20% - 30% > 30% PC Diesel < 5% 5% - 30% > 30% LDV Petrol < 40% 40% - 60% > 60% LDV Diesel < 5% 5% - 25% > 25% HGV Diesel < 5% 5% - 40% > 40% Technology standard A is the standard referred to for the European (and North American) Fleet of vehicles. Technology standards B and C are dependent on the time delay in implementation of vehicle emission standards for specific countries (i.e. 10 years for standard B and no effective emission controls/programs for standard C). There are a few potential limitations with the simplified method from PIARC (2012). 1. Future Influencing Factors Post 2020 Since the technology standard is determined by the percentage of Pre-Euro vehicles, the introduction of later Euro standards and their uptake are not taken into consideration. This means the Australian Fleet is currently forecasted until 2020 based on this assumption. The future year influencing factors (fe) are provided for the years 2015 and No information is provided in PIARC (2012) for years in-between 2015 and 2020, nor is there any information on factors for years up to and including Australia-wide Emission Factors Emission factors are provided for Australia as a nation and may not necessarily reflect the Fleet within NSW and trends in fuel preferences. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

11 2 NSW Fleet Analysis 2.1 Overview The NSW Fleet is characterised by the age of each vehicle registered within NSW. The vehicle emission standard documents for Australia are referred to as the Australian Design Rules for (ADR). Each ADR defines the exhaust emission limits for CO, NO x and PM, and are generally aligned with European standards (Euro) in terms of the specified limits. The implementation dates of the Australian standards, however, differ from those in Europe. In order to reconcile this difference, this report analyses the age distribution in each year for NSW between 2000 and 2015, and uses the trends for each vehicle class to forecast the registered Fleet to The NSW Fleet forecasted by this work adds a further refinement by weighting the Fleet based on vehicle kilometres travelled. In doing so, the Fleet is weighted to reflect the relatively greater distance that newer vehicles travel, in comparison to older vehicles. Thus, the methodology for the Fleet forecast has been separated into two stages: Stage 1: Forecast of the Fleet based on registration data for vehicles in NSW Stage 2: Adjustment of the NSW Fleet for vehicle age and vehicle kilometres travelled (VKT) DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

12 3 Stage 1: NSW Registration Fleet Forecast In order to gain an understanding for the process and methodology used to analyse the NSW Fleet, the case of passenger vehicles will be used for this section. Figure 1 presents an outline of the steps involved in this process. Figure 1: Outline of methodology to determine the NSW Registration Fleet Forecast Part A Information Gathering Part B Forecasting Methodology Part C Alignment with Emissions Standards NSW Registration Fleet Forecast 3.1 Part A: Information Gathering The information used for Stage 1 of this process was obtained from the publicly available RMS registration statistics website: The following three data sets, available from the RMS registration statistics website, have been utilised for calendar year analysis from 2000 to 2015: Year of manufacture by vehicle type for registered vehicles at each year Motive Power by vehicle type for registered vehicles at each year New registrations by fuel type by vehicle type at each year Data set contains, for each past year, the number of vehicles that exist in the NSW Fleet for a given year from a certain year of manufacture. Data from 2015 is used in this section to demonstrate the methodology. The data used is taken from the 4 th quarter of each year as this data set reflects the full calendar year. Data sets and provide data that informs the percentage split between petrol and diesel fuel types for the different vehicle classes. They provide information for both the overall aggregated Fleet and all new vehicles registered in a given year. For data set 1.1.4, the data was taken from the end of each calendar year. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

13 By combining the three data sets (1.1.3, and 1.2.6), the number of vehicles can be determined for any combination of the following three categories: Year of Manufacture Fuel Type (Petrol/Diesel) Vehicle Class (PC, LDV, HGV) As an example, it would be possible to determine the number of Petrol Passenger registered in 2004 or the number of Diesel Light Duty registered in the year The reason for separating vehicles into their classes and fuel types is in order to align them with the PIARC (2012) vehicle categories. These categories are: Petrol Passenger Diesel Passenger Petrol Light Duty Diesel Light Duty Heavy Goods (Assumed to be 100% Diesel post 2015) A discussion on the classification of vehicles and fuel types, as used in this report, is provided in section The following tables present examples of the information data sets available from the RMS Registration Statistics Website. The percentage values presented in Table 7 and Table 8 have been consolidated from Data sets and respectively as the overall split between petrol and diesel for each vehicle class. Tables 7 and 8 contain the percentage splits between petrol and diesel vehicles from 2000 to DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

14 Table 4: Number of passenger vehicles from each Year of Manufacture, 2015 (Example) Year of Manufacture Total Number of Passenger TOTAL: Consolidated data obtained from the RMS website: Year of manufacture by vehicle type for registered vehicles as at 2015, quarter 4. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

15 Table 5: Consolidated 2015 Registration Data, Registrations by Fuel type by Vehicle type (Example) Petrol Diesel Natural Gas Hydrogen/ Electric 2015 Passenger Light Duty Heavy Fuel Type Passenger Off-road People movers Small Buses Mobile Homes Light Trucks Buses Heavy Trucks Prime Movers Unleaded Petrol 2,689, ,469 59,818 7,419 4, , ,866,287 Petrol 49,577 2, , ,903 LPG 15,602 1, , ,451 Petrol And LPG (Dual Fuel) 22,535 15, , ,801 Petrol & Compressed Natural Gas Diesel 116, ,154 8,769 9,509 5, ,346 12,486 92,533 20,953 1,013,236 Diesel And LPG (Dual Fuel) Diesel & LPT (Torque Topping) Diesel/NAT Compressed Natural Gas Liquid Natural Gas Hydrogen Electric/Petrol 17,578 2, ,001 Electricity Total Total Petrol 3,727, ,975 1,902 Total Diesel 505, , ,316 TOTAL VEHICLES 4,254, , ,872 Percentage of Petrol 87.63% 42.92% 1.48% Percentage of Diesel 11.89% 57.02% 98.02% Consolidated data obtained from the RMS website, Motive power by vehicle type registered vehicles as at 2015, quarter 4 DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

16 Table 6: Consolidated 2015 Registration Data, New Registrations by Fuel type by Vehicle type (Example) 2015 Passenger Light Duty Heavy Fuel Type Passenger Off-road People movers Small Buses Mobile Homes Light Trucks Buses Heavy Trucks Prime Movers Total Unleaded Petrol 175,433 75,288 2, , ,370 Petrol Petrol Diesel Natural Gas Hydrogen/ Electric LPG Petrol And LPG (Dual Fuel) Petrol & Compressed Natural Gas Diesel 14,567 40, , , ,233 Diesel And LPG (Dual Fuel) Diesel & LPT (Torque Topping) Diesel/NAT Compressed Natural Gas Liquid Natural Gas Hydrogen Electric/Petrol 1, ,784 Electricity Total Petrol 253,055 5,767 9 Total Diesel 55,686 49,733 3,870 TOTAL VEHICLES 310,723 55,616 3,881 Percentage of Petrol 81.44% 10.37% 0.23% Percentage of Diesel 17.92% 89.42% 99.72% Consolidated data obtained from the RMS website, New registrations by fuel type by vehicle type for 2015 DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

17 Table 7: Percentage split (Petrol/Diesel) for all vehicles registered, from 2000 to 2015 Year Passenger Petrol Light Duty Heavy Goods Passenger Diesel Light Duty Heavy Goods % 74.70% 6.77% 3.04% 25.28% 93.02% % 74.38% 6.29% 3.25% 25.61% 93.41% % 73.77% 5.92% 3.47% 26.21% 93.71% % 73.14% 5.44% 3.67% 26.85% 94.18% % 72.31% 4.95% 3.85% 27.67% 94.67% % 71.20% 4.37% 4.01% 28.79% 95.27% % 70.01% 4.08% 4.26% 29.98% 95.56% % 68.53% 3.74% 4.66% 31.46% 95.90% % 66.57% 3.45% 5.23% 33.42% 96.18% % 64.43% 3.26% 5.81% 35.55% 96.25% % 61.87% 3.03% 6.63% 38.10% 96.46% % 59.05% 2.83% 7.45% 40.91% 96.63% % 56.03% 2.47% 8.43% 43.94% 96.98% % 52.20% 1.94% 9.49% 47.76% 97.51% % 48.57% 1.74% 10.50% 51.39% 97.72% % 42.92% 1.48% 11.89% 57.02% 98.02% Table 8: Percentage split (Petrol/Diesel) for newly registered vehicles each year, from 2000 to 2015 Year Passenger Petrol Light Duty Heavy Goods Passenger Diesel Light Duty Heavy Goods % 74.70% 6.77% 3.04% 25.28% 93.02% % 73.92% 3.06% 4.42% 26.07% 94.67% % 69.66% 2.62% 4.71% 30.33% 96.22% % 67.65% 0.57% 4.39% 32.34% 99.38% % 65.45% 0.60% 4.64% 34.40% 99.28% % 60.49% 1.20% 5.28% 39.50% 98.70% % 53.71% 0.84% 7.13% 46.28% 99.07% % 45.70% 0.59% 9.82% 54.29% 99.41% % 41.35% 0.84% 12.55% 58.63% 97.83% % 34.63% 0.57% 14.01% 65.23% 97.96% % 29.10% 1.12% 16.52% 70.81% 97.57% % 25.86% 0.90% 17.19% 74.01% 98.58% % 18.62% 0.87% 19.48% 81.32% 98.81% % 13.35% 0.59% 19.67% 86.63% 99.29% % 11.31% 0.56% 19.07% 88.61% 99.13% % 10.37% 0.23% 17.92% 89.42% 99.72% DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

18 The percentage values from Table 8 are applied to the number of vehicles from the respective year of manufacture. This is presented in Table 9, below. This step makes the assumption that the percentage split between petrol and diesel remains constant for vehicles from a particular year of manufacture, for all subsequent years. For example, if the percentage of new diesel fuelled vehicles registered in 2003 was 4.39% then it is assumed that 4.39% is the percentage of 2003 year model vehicles that are diesel fuelled in 2004, 2005, etc. Table 9: Determination of the number of petrol/diesel vehicles, given the percentage of new diesel vehicles registered during the year of manufacture Year of Manufacture Total Number of Passenger Percentage of New Diesel Petrol Passenger Diesel Passenger % % % % % % % % % % % % % % % % % % % % DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

19 3.1.1 Discussion on Fuel and Vehicle Type Classifications Fuel Type: Classifications and Observations Classification of Fuel Types The following classifications have been used to define the main different fuel types in the analysis of registration data from data sets and Petrol: Unleaded Petrol Petrol LPG Petrol and LPG (Dual Fuel) Petrol & Compressed Natural Gas Diesel Diesel Diesel and LPG (Dual Fuel) Diesel & LPT (Torque Topping) Diesel/NAT Alternative Low Emission Fuels The alternative, low emission fuels within this discussion are considered to be Natural Gas, Hydrogen, Electric and Hybrid vehicles. For the 2015 NSW Fleet, alternative, low emission fuels currently represent a small percentage of vehicles at 0.38%. Since 2005, the percentage of new alternative, low emission fuels registered each year has increased, with a spike in 2012 of 1%. Due to the negligible proportion that alternative, low emission fuels represent in the vehicle Fleet, they have been excluded from the analysis of the percentage distribution of fuel types. Table 10 presents the percentage of existing alternative, low emission fuelled vehicles in the NSW Fleet between 2005 and Table 10: Percentage of Alternative low emission fuels between 2005 and 2015 for the NSW Fleet Percentage of Alternative Fuelled % 0.03% 0.04% 0.05% 0.07% 0.11% 0.15% 0.19% 0.32% 0.34% 0.38% Consolidated data obtained from the RMS website: Motive Power by vehicle type for registered vehicles at each year DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

20 Observations regarding the fuel split between petrol and diesel When analysing the split between petrol and diesel for passenger vehicles, LDVs and HGVs, a general trend is observed whereby the percentage of petrol vehicles is decreasing while the percentage of diesel vehicles is increasing. Table 11 presents this trend, taking into account that heavy goods vehicles are estimated to be 100% diesel fuelled from 2015 onwards. In comparing the PIARC detailed method to the PIARC simple method, there is a significant difference in the fuel split between petrol and diesel for Passenger Cars (PCs) and Light-Duty (LDVs). The detailed method emission factors for PCs and LDVs (presented in Appendix A of this report) utilises a fuel mix of petrol and diesel, instead of separate individual factors for petrol and diesel. That is, for any given year, the emission factors reflect the mix of petrol and diesel fuelled PCs and LDVs within the Fleet. The Australian simple method fuel split is significantly different to those of Technology Standard A in Europe. PIARC (2012) states that 96% of Light Duty in Technology Standard A are diesel fuelled. Table 12 contains the fuel split ratios for Australia and Europe (simple methods) as stated in PIARC (2012) for the period from 2010 to The assumption made in PIARC (2012), a 50-50% distribution between petrol and diesel LDVs for 2010, is different from the 38.1% for diesel fuelled LDVs recorded by RMS registration statistics in As the percentage of petrol and diesel LDVs differs significantly from PIARC (2012) data, further refinement by LDV fuel type would improve the accuracy of emission factors for LDVs within the NSW Fleet. For 2030, the percentage of diesel fuelled LDVs is forecasted to be 80.70%. For the emission factors calculated in Appendix A, the overall increase in the percentage of diesel fuelled vehicles was taken into consideration. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

21 Table 11: Percentage distribution between petrol and diesel from 2000 to 2040 Petrol Diesel Year Passenger Light Duty Heavy Goods Passenger Light Duty Heavy Goods % 74.70% 6.77% 3.04% 25.28% 93.02% % 71.20% 4.37% 4.01% 28.79% 95.27% % 61.87% 3.03% 6.63% 38.10% 96.46% % 42.92% 0.00% 11.89% 57.02% 100% % 39.33% 0.00% 16.20% 60.67% 100% % 19.30% 0.00% 25.21% 80.70% 100% % 0% 0.00% 34.22% 100% 100% Table 12: Fuel Mix of LDVs (average ) in Europe and Australia. Information obtained from PIARC (2012) - Table 13, p28 and Table 36, p42 Fuel Type Europe (Technology Standard A*) Australia Petrol 4% 50% Diesel 96% 50% *Technology Standard A is defined within PIARC (2012). The standard is presented as a relative proportion of vehicles for the base year 2010 under each Euro emission standard. Technology Standard A covers the European Union, USA and countries with similar vehicle emission standards and years of implementation. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

22 Vehicle Types: Classifications Classification of Vehicle Types The following classifications have been used to define the main different fuel types in the analysis of registration data from data sets 1.1.3, 1.1.4, and Passenger Light Duty Heavy Goods Passenger vehicles Small Buses Buses Off-road vehicles Mobile Homes Heavy Trucks People movers Light Trucks Prime Movers Motorcycles and Scooters The categories of motorcycles and scooters have been omitted from the analysis of registration data within this report. This assumption was made on the basis that the vehicle classes outlined in PIARC (2012) do not characterise motorcycles as passenger vehicles, nor do they provide a separate vehicle class for motorcycles or scooters. If motorcycles were included in the data set, the percentage of petrol passenger vehicles would increase and the percentage of diesel passenger vehicles would decrease. Table 13 highlights the minor differences in the percentage distribution forecasts for petrol and diesel passenger vehicles. Therefore, having excluded motorcycles and scooters from this analysis, the resultant percentage of diesel vehicles for 2020 and 2030 is greater than would be the case if motorcycles and scooters were included. Table 13: Forecasted percentage distribution for petrol and diesel until 2030 with and without the inclusion of motorcycles and scooters in the passenger vehicle mix Fuel Type Passenger vehicles (including motorcycles and scooters) Petrol 96.11% 93.53% 88.93% 85.68% 82.09% 78.50% Diesel 3.88% 6.34% 10.63% 13.70% 17.07% 20.45% Passenger vehicles (omitting motorcycles and scooters) Petrol 95.98% 93.25% 88.37% 84.69% 80.66% 76.63% Diesel 4.01% 6.63% 11.18% 14.64% 18.43% 22.22% Consolidated data obtained from the RMS website: Motive Power by vehicle type for registered vehicles at each year DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

23 3.2 Part B: Registration Fleet Forecasting Methodology Having aligned the NSW Registration Data with the PIARC (2012) vehicle categories (i.e. Petrol Passenger Cars), it is possible to analyse the growth and decay trends for each individual category, and to forecast how they might continue into the future. For an individual category, in any given year, the number of vehicles from each year has already been determined, as outlined in Table 9. The number of registered vehicles follows a year-on-year trend whereby it increases slightly from year one to year two, and thereafter gradually decreases. An example of this is presented in Figure 2, below, for petrol passenger vehicles manufactured in Figure 2: Change in the number of registered Petrol Passenger manufactured in 2003 By calculating the year-on-year change as a percentage, the growth/decay trend can be determined for a unique year model vehicle. It has been observed that the trend in Figure 2 is consistent for all unique year model vehicles. The year-on-year percentage changes are presented in Table 14 for all years from 2005 to Figure 3 to Figure 7 present the trends in the total number of vehicles for each PIARC category between 2000 and Figure 8 presents the growth in the NSW Fleet as a stacked graph from 2000 to 2015, highlighting the linearly increasing historic trend of the Fleet as a whole. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

24 Figure 3: Total Number of registered Petrol Passenger 2000 to 2015 Figure 4: Total Number of registered Diesel Passenger 2000 to 2015 Figure 5: Total Number of registered Petrol Light Duty 2000 to 2015 DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

25 Figure 6: Total Number of registered Diesel Light Duty 2000 to 2015 Figure 7: Total Number of registered Heavy Goods 2000 to 2015 Figure 8: Overall Growth of the NSW Registered Fleet 2000 to 2015 DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

26 Table 14: Year-on-year change, represented as a percentage for 2006 to 2015 Year of Manufacture % % % % 99.94% % % 99.05% % % % 99.20% 99.51% % % % 98.67% 99.10% 99.14% % % % 98.17% 98.28% 98.75% 98.86% 99.49% % % 98.10% 98.33% 98.57% 98.93% 98.85% % % % 97.86% 98.26% 98.51% 98.72% 99.12% 98.97% % % % 97.00% 97.58% 97.95% 98.06% 98.29% 98.66% 98.92% % % Beyond 2015, the forecasts are based on the growth and decay trends as seen in Figure 2, depending on the unique year model vehicle. Table 15 presents how the year-on-year trend is applied. A vehicle that is two years old will follow a similar growth/decay trend for 2015 (100.2%) as it would in 2016 (100.31%), in 2017 (100.26%) and so on. The blue highlighted cells in Table 15 represent the year-on-year change for a vehicle that is two years old. The value for each blue cell is calculated by taking the average of the blue cells from previous years (as seen in the sample calculation below for 2 year old vehicles in 2022) % = 100.1% % % % % % % % 8 Table 15: Year-on-year change (forecast), represented as a percentage for 2013 to 2022 Year of Manufacture % % % % % % % 98.57% % % % 98.44% 98.62% % % % 98.36% 98.49% 98.63% % % % 98.19% 98.43% 98.57% 98.66% % % % 98.04% 98.28% 98.48% 98.67% 98.62% % % % 97.66% 98.11% 98.29% 98.50% 98.73% 98.68% % % % 97.23% 97.77% 98.15% 98.34% 98.55% 98.81% 98.60% % % DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

27 The year-on-year percentage changes were translated back into absolute values for the years to show the growth/decay between years in the units number of vehicles. Table 16 presents the absolute number of petrol passenger vehicles forecasted for 2016 to Table 16: Number of petrol passenger vehicles (forecast) for 2016 to 2022 Year of Manufacture , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , , ,713 This forecast methodology was applied to all future years from 2016 up until DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

28 3.3 Part C: Alignment with Vehicle Emission Standards The calculations for vehicle emissions, as presented in PIARC (2012), are based on vehicle emission standards. Emission standards mandate the maximum allowable rate of tailpipe emissions for different pollutants that all new vehicles must comply with. In Australia vehicle emissions standards defined in the applicable Australian Design Rule (ADR) and are equivalent to the Euro standards. The assumed mandated implementation years for each emission standard are presented in Table 17. In order to align the RMS registration data and forecasts with the vehicle emission standards, the number of vehicles from each year of manufacture must be represented as a proportion of the total number of vehicles in each individual category. This data can then be presented for any year of interest (i.e. a future design year). Table 17: Periods of implementation for vehicle emission standards in Australia There is a time lag of a number of years between the first international implementation of European vehicle emission standards and their implementation as part of the Australian Design Rules. Table 18a and Table 18b demonstrate this time lag between implementing corresponding European and Australian vehicle emissions standards. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

29 Table 18a: Comparison of dates of implementation for vehicle emissions standards between Europe and Australia Petrol Passenger and LDVs Diesel Passenger and LDVs Standard Europe Australia Europe Australia Euro 1 (ADR37/01) # # Euro 2 (ADR79/00) Euro 3 (ADR79/01) (not adopted) Euro 4 (ADR79/02) Euro 5a (ADR79/03) Euro 5b (ADR79/04) * * Euro 6 (ADR79/05) * * Table 18b: Comparison of dates of implementation for vehicle emissions standards between Europe and Australia Heavy Goods Standard Europe Australia Euro I (ADR70/00) Euro III (ADR80/00) Euro IV (ADR80/02) Euro V (ADR80/03) Euro VI 2013 (not proposed) *Proposed implementation date # The Australian implementation dates are an estimate based on equivalent standard definitions, e.g. CO and NOx levels, but were not actual corresponding Euro 1 standards for the Australian Fleet. This was done for the purpose of analysis and comparison. Note1: The dates provided are the final year of the implementation period whereby all new vehicles post and including this year must comply with the standard. Note 2: The standard compared for Light Duty is Euro Category N 1 Class III. Data obtained from and DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

30 3.4 Conclusion The process of aligning NSW vehicle registration data with PIARC (2012) categories was undertaken for passenger vehicles, light-duty vehicles and heavy goods vehicles for each year from 2000 to Table 19 presents the NSW registration-based Fleet for Table 19: NSW Registered Fleet, 2015 Type Year Pre ADR(37/01) ADR(37/01) ADR(79/00) ADR(79/01) ADR(79/02) ADR(79/03) PC gasoline 8.30% 24.47% 10.69% 21.67% 23.09% 11.78% PC Diesel 2.11% 6.01% 14.81% 55.29% 21.78% LDV Gasoline % 31.44% 12.61% 17.50% 11.75% 3.46% LDV Diesel 6.42% 7.45% 16.02% 46.92% 23.18% HGV Diesel 24.66% 18.19% 23.44% 17.04% 16.67% DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

31 4 Stage 2: NSW VKT-weighted Fleet Forecast This section describes the process by which a vehicle-kilometres-travelled (VKT) weighting is applied to the NSW registration-based Fleet, relative to the age of the vehicle. Forecasting for passenger vehicles will be used as an example in this section as it was for the analysis of the NSW registration-based Fleet presented in Section 3. Figure 9 presents an outline of the steps involved in this process. Figure 9: Outline of the methodology used to forecast the NSW Fleet from 2016 to 2040 Part A Information Gathering Part B VKT-weighting Methodology NSW VKT-weighted Fleet Forecast 4.1 Part A: Information Gathering The information used for Stage 2 of this process was obtained from the following three data sets, available at the publically available ABS website (below) Survey of Motor Vehicle Use, Australia, 12 months to 31 October Survey of Motor Vehicle Use, Australia, 12 months to 31 October Survey of Motor Vehicle Use, Australia, 12 months to 31 October The relevant table from the above data sets was titled Total kilometres travelled, by year of manufacture by type of vehicle: by state/territory of registration This data provides a breakdown of the number of kilometres travelled by vehicles based on their: Year of Manufacture Vehicle Class (PC, LDV, HGV) State of registration What the information does not provide is the difference in VKT by petrol or diesel vehicles. Therefore, this methodology assumes that petrol and diesel vehicles, of a certain age, drive equal distances. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

32 The year of manufacture information was presented as three distinct blocks, as seen in Table 20. The three block periods are of equal size, but the bands of years (i.e or ) are different for the three data sets. Each statistic (the number of kilometres driven) is accompanied by a Relative Standard Error (RSE %). This RSE percentage, when doubled and applied to the number of kilometres travelled, gives the maximum and minimum number of kilometres travelled by 95% of the vehicles in that category. This is presented in Figure 10. Table 20: Estimated Vehicle Kilometres Travelled in NSW for vehicle classes and age NSW RSE NSW RSE NSW RSE (million) (%) (million) (%) (million) (%) 16 + Years Old 1994 and earlier 1996 and earlier 1998 and earlier Passenger vehicles 6, , , Light commercial vehicles 1, , Rigid trucks Years Old 1995 to to to 2008 Passenger vehicles 21, , , Light commercial vehicles 4, , , Rigid trucks , , to 5 Years Old 2005 and after 2007 and after 2009 and after Passenger vehicles 21, , , Light commercial vehicles 5, , , Rigid trucks 1, , DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

33 4.2 Part B: VKT Weighted Fleet Forecasting Methodology To forecast the Fleet and weight it for vehicle kilometres travelled (VKT), an age profile of the Fleet is required. The Fleet age profile informs the likelihood that a vehicle will be driving on the road, depending on its age. The vehicle-kilometrestravelled weighting can then be applied to the age profile. The data presented in Table 20 shows that newer vehicles travel further per year than older vehicles. The format of the data sets, however, poses a limitation. It is not possible to obtain a breakdown of the years of manufacture from the Australian Bureau of Statistics as the associated RSEs (>50%) are too great for the data to be reliable. An RSE of 50% means that the range of VKT travelled lies between plus or minus 100% of the value provided. As such, the split of VKT for each block period was done to reflect the registration statistics within the respective period. For example, to distribute the VKT by passenger vehicles 0 to 5 years old in 2014, the following split was applied (see Table 21). The number of registered vehicles data in Table 21 is obtained from the NSW registration data set of Table 21: Distribution of VKT within a block period (example): 0-5 years old, 2014 Year of Manufacture Relative Age (Years) Number of registered vehicles Relative VKT within (0-5 yr) block period Relative VKT as % , % , % , % , % , % , % Total VKT: 21,865 This process was repeated for all block periods in the three data sets. A weighting was done to show the relative VKT as a percentage of all the years of manufacture. The RSEs were then applied and an average of the three data sets was taken. The result of this is visually presented in Figure 10 for the kilometres travelled by passenger vehicles in 2014 and Figure 11, for the average VKT age profile. This process was repeated from 2000 to 2040 and applied to the NSW Registration Fleet Forecast (from Stage 1, Section 3) to obtain a vehicle kilometre weighted Fleet. DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

34 Figure 10: Candlestick graph of Passenger VKT (2014) with upper and lower bounds of the RSE Figure 11: Average Passenger Vehicle Kilometres Travelled Age Profile In order to present the data set as a forecast that can subsequently be readily used, understood and applied, each vehicle class has been represented as 100% (see Table 22). Thus, the split between petrol and diesel is incorporated into the results without the need for a separate table. In short, when given the total number of passenger vehicles, it is possible, from Table 22, to determine the number of Diesel Euro 4 Passenger by just applying the appropriate percentage. Table 22: NSW VKT-weighted Fleet, 2015 Vehicle Fuel Year Pre Euro Euro 1 Euro 2 Euro 3 Euro 4 Euro 5 Euro 6 Total PC Petrol 8.26% 12.49% 8.32% 19.41% 29.25% 11.25% 0.00% Diesel 0.22% 0.31% 1.28% 0.00% 6.97% 2.23% 0.00% 100% LDV Petrol % 6.98% 4.84% 7.89% 7.23% 1.82% 0.00% Diesel 4.71% 1.58% 6.73% 0.00% 29.43% 12.60% 0.00% 100% HDV Diesel 35.49% 6.75% 0.00% 16.04% 20.62% 21.10% 0.00% 100% DRAFT NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

35 4.3 NSW Fleet Forecast for Ventilation Design The process set out in section 4 for passenger vehicles was then repeated for light-duty vehicles and heavy goods vehicles for each year from 2000 to Figures 12 to 14 present the historic and forecasted percentage of vehicle kilometres travelled by each vehicle type and emission standard in a year. Figure 12: Percentage of all kilometres travelled by Passenger in NSW NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

36 Figure 13: Percentage of all kilometres travelled by Light Duty in NSW NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

37 Figure 14: Percentage of all kilometres travelled by Heavy Goods in NSW NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

38 5 References Department of Infrastructure and Regional Development, Vehicle Emission Standards. [Online] Available at: [Accessed 3 July 2015]. Hausberger, S., Rexeis, M., Zallinger, M. & Luz, R., Emission Factors from the Model PHEM for the HBEFA Version 3, Bern: HBFEA. NSW EPA, Air Emissions Inventory for the Greater Metropolitan Region in New South Wales Calendar Year. Technical Report No. 7 - On-Road Mobile Emissions: Results, Sydney South: NSW Environment Protection Authority. Pastramas, N., Samaras, C., Mellios, G. & Ntziachristos, L., Update of the Air Emissions Inventory Guidebook - Road Transport 2014 Update, Thessaloniki: EMISIA SA. PIARC, Road Tunnels: Vehicle Emissions and Air Demand for Ventilation, La Defense cedex: World Road Association (PIARC). Roads and Maritime Services, Registration. [Online] Available at: [Accessed 9 July 2015]. Transport Policy.net, Emissions Standards. [Online] Available at: [Accessed 3 July 2015]. NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to

39 Appendix A Case Study: NSW Emission Factors 2015 A.1 Overview This appendix presents the case of applying the NSW Fleet to the emission factors as part of the PIARC (2012) detailed method. Having determined the NSW VKT-weighted Fleet in Section 4, the raw data presented in PIARC (2012), together with the Fleet forecast described here, can be utilised to calculate emissions specific to NSW. This raw data presented in PIARC (2012), referring to each individual Euro-standard, was derived from the Handbook of Emission Factors for Road Transport (HBEFA), version 3.1. (Hausberger, et al., 2009). The methodology for the determination of these factors can be found in the publicly available document through the link below: The HBEFA utilises the Passenger Car and Heavy Duty Emission Model (PHEM) as the basis for their calculated emissions. For the purposes of this report, it is assumed that the raw data calculated and published for individual Euro-standards by HBEFA and moreover, by PIARC (2012), are accurate and can be applied to the NSW Fleet. The raw base data in PIARC (2012) is presented in tables in section 4 of their report for each vehicle category, exhaust pollutant and gradient. Table 25 is an example of the information contained in one such table for CO emissions (grams/hour) from petrol passenger vehicles along a gradient of 0%. In this table, the PIARC (2012) degradation factors for Petrol Passenger vehicles have already been applied. The application of the degradation factors are explained in section A.2. NSW Fleet Forecast for Tunnel Ventilation Design: 2016 to