M5 East Motorway tunnels and air quality issues

J.R. Anderson Roads & Traffic Authority, NSW

ABSTRACT: Forming part of the Sydney Orbital road network, the 10 km long M5 East motorway extends from Beverly Hills to General Holmes Drive at Sydney Airport. The new toll free M5 East motorway includes two parallel, unidirectional two-lane tunnels each 4 km long. The ventilation system includes a single 35m high stack connected to near the midpoint of the mainline tunnels by a 700-metre long ventilation tunnel. Delivery of the project is by a design and construct contract with an operate and maintain period of ten years. This paper briefly reviews the history of development of the project and the project approval conditions. The paper also reviews, for the approved project, air quality issues associated with the project, community involvement during development of the design, the air quality monitoring program, background air quality monitoring and provision for air cleaning systems. 1 INTRODUCTION The M5 East freeway forms part of the Sydney Orbital road network and will provide a new, untolled link between the M5 motorway and the Eastern Distributor. The M5 East is a major State Government and the Roads & Traffic Authority NSW (RTA) initiative to improve road safety and traffic flow in southwestern Sydney. The four-lane divided dual carriageway freeway is ten kilometres in length, including a 4 kilometre long tunnel between Bexley Road at North Bexley and Marsh Street near Sydney Airport. This paper gives an overview of the development of the approved project and outlines details of the tunnel ventilation system and associated air quality issues. 2 PLANNING HISTORY Provision for a road from the centre of Sydney west to Liverpool has been made in many planning documents and studies in the past. A County Road Reservation along the Wolli Creek Valley was included in the County of Cumberland Planning Scheme gazetted in 1951. In recent times an Environmental Impact Statement (EIS) was exhibited in 1989. The proposal included a six-lane elevated structure in the County Road Reservation along the Wolli Creek valley, east of Bexley Road. No determination was made on the proposal. The change from a surface road along the Wolli Creek valley to a tunnel also provided a significant environmental benefit in retaining bushland (to be managed as a Regional Park) for the surrounding communities of Earlwood and Turrella. The 2.5-km long tunnel proposal in the 1994 EIS included a 15-metre high exhaust stack, located on the adjacent ridge in the Earlwood shopping centre car park approximately 220 metres north of the tunnels. Air intake was from the portals. A 3.6-km extended tunnel option was also fully described. This option included an additional 30m Subsequently a further EIS was exhibited in 1994 for a tolled M5 East motorway between King Georges Road and General Holmes Drive. The most significant changes between this proposal and the one in the 1989 EIS were:

four lanes of traffic instead of six a 2.5km tunnel under Wolli Creek on the existing road reserve east of Bexley Road to near Turrella Railway Station no toll

The downsizing of the capacity of the motorway in the 1994 EIS reflected an integrated approach to transport planning with emphasis on an appropriate balance between private vehicle usage and public transport systems. The M5 East also has a key objective of supporting economic development by facilitating freight movements and reducing traffic levels (particularly heavy transport vehicles) in residential streets.

high ventilation stack in the Wolli Creek valley adjacent to Turrella Railway Station near the current stack location. Subsequently, in response to comment received from the community, a supplement to the 1994 EIS was exhibited in 1996 proposing a 4km direct tunnel from a portal at Bexley Road to a new portal at Marsh Street near Sydney Airport. The proposed tunnels included three, 15 metre high exhaust stacks, located in residential areas on ridgelines above the tunnels. Four air intake locations, apart from the portals, were also proposed. Following exhibition of the EIS, in response to further community comment a number of changes were made to the proposal in June 1997 to further reduce environmental impact. These changes included replacing the three stacks in residential area with a single stack in an industrial area in the County Road Reservation at Turrella as shown in Figure 1.

The conditions also require:

the tunnel ventilation system to be designed to avoid emissions from the portals as far as practical a minimum stack height of 25 metres with final stack height to be approved by the Director General of the Department of Urban Affairs & Plann ing (DUAP). wind tunnel testing of the stack provision for installation of air treatment systems including Electrostatic Precipitators which can only be required if there is an exceedance of air quality goals due to the stack a comprehensive air quality monitoring system establishment of an Air Quality Community Consultative Committee

Figure 1. Approved Alignment 3 PROJECT APPROVAL

The Minister for Urban Affairs & Planning approved the proposed M5 East on 9 December 1997, subject to 150 conditions. A number of these conditions require the ventilation design and corresponding vehicle emissions not to exceed nominated air quality goals in the tunnels and not to exceed nominated ambient air quality goals at ground level.

preparation of a Subregional Air Quality Management Plan

Since project approval opponents of the project have unsuccessfully sought recourse to the Land and Environment Court. A Legislative Council (Upper House of NSW Government) Committee has also held two inquiries (1999 and 2001) into the ventilation stack.

4 ECONOMIC IMPLICATIONS The road reservation for what has now become the M5 East project has been in place for over 50 years. In response to community input, the proposed motorway between Bexley Road, Earlwood and Marsh Street, near Sydney Airport, has changed, through the course of the M5 East assessment and approval process, from a surface motorway in a gazetted corridor in the Wolli Creek Valley to include a 4km tunnel in an entirely new corridor. Each of the changes has increased the capital cost of the M5 East to the current capital cost of $794 million. The cost of ventilating the tunnel has also progressively increased to reflect the increasing length of the tunnel, progressively more stringent air quality goals (eg. the ambient air quality goal for PM 10 has reduced by a factor of three from 150 g/m3 t o 5 0g/m3), and other strict approval conditions (including the requirement for no emissions from the portals). Operating costs of the M5 East have similarly increased as the tunnel length and other tunnel constraints have changed to meet new requirements. While energy needs for dealing with tunnel emissions have increased they represent a small proportion of the total increase in cost. Although the capital cost of the project has increased from changes to the project made in response to community comment, the benefit:cost ratio is in excess of 2.0, making the M5 East economically viable. This is a result of benefits from improved traffic flow on the road network. There are also other benefits to the community from the removal of traffic (including heavy truck traffic) from many local streets, providing increased safety and amenity for residents. 5 ACCEPTANCE OF TENDER Tenders for design, construction and operation and maintenance of the motorway for a period of ten years were invited from three short listed contractors in July 1997. On 14 August 1998 a contract was awarded to the Baulderstone Hornibrook Bilfinger Berger Joint Venture (BHBB) for design and construction as well as operation and maintenance of the freeway for 10 years.

6 TUNNEL VENTILATION The 4km parallel unidirectional tunnels have a longitudinal ventilation system modified to meet the approved project constraints including a single exhaust stack at Turrella and no emission (as far as practical) from the portals. Jet fans are provided over the tunnel length with axial fans installed in cross over passages adjacent to the portals to move emissions to the adjacent tunnel to avoid emission at the portals. Emissions are extracted from each tunnel near the midpoint via the 700m ventilation tunnel to the 35m high exhaust stack at Turrella. Eight axial fans (at least one always on standby) are located at Turrella. Fresh air also enters the tunnel near the midpoint via a 70m vertical shaft from an air intake station at Duff Street. A sketch of the ventilation system designed by Hyder Consulting Australia Pty Ltd, is given in Figure 2. 7 AIR QUALITY 7.1. 7.1.1 Air Quality Goals In-Tunnel

The conditions of approval require that the tunnel be operated so that a 15-minute time weighted average goal for Carbon Monoxide of 87 ppm is not exceeded during operation at any location. The project deed also requires that visibility meet the PIARC criteria of an extinction factor of 0.005/metre (free flowing traffic) or 0.007/metre (congested traffic). 7.1.2 External

The tunnel exhaust stack must be designed so that emissions do not result in ambient air quality at ground level exceeding: NO2 One hour average of 256 g/m3 PM 10 24 hour average of 50 g/m3

LEGEND TURRELLA EXHAUST STACK MAIN LINE DRIVEN TUNNEL JET FAN AIR FLOW

EA

ST

BO

UN

VENTILATION EXHAUST TUNNEL

TU

NN

WE

EL

ST

BO

DUFF STREET SUPPLY VENTILATION STATION

UN

TU

NN

EL

PRINCES HIGHWAY EXIT RAMP

MARSH STREET MARSH STREET EXIT RAMP EXIT RAMP

MARSH STREET ENTRY RAMP

Figure 2. Tunnel Ventilation System The design must have regard for the current short term Victoria EPA 3-minute design ground level concentration criteria of: Toxic Organic Compounds Benzene 3 minute average of 0.1 mg/m3 1-3 Butadiene 3 minute average of 1.0 mg/m3 Odorous Compounds Acetaldehyde 3 minute average of 0.076 mg/m3 Formaldehyde 3 minute average of 0.10 mg/m3 7.2 7.2.1 Air Dispersion Modelling General dispersion consultant (Air Noise Environment) undertook both numerical and physical dispersion modelling in order to demonstrate compliance of the stack emissions with the project ambient air quality goals (Hyder 2000). 7.2.2 Numerical Dispersion Modelling

The overall objective of the numerical dispersion modelling was to confirm that emissions from the tunnel were appropriately dispersed in the atmosphere in order to meet the project air quality goals. Following input from the community a decision was made to undertake the numerical modelling for stack heights of both 25 and 35 metres. Ambient air quality meteorological data was sourced from the Environment Protection Authoritys (EPA) Earlwood (Beaman Park) monitoring statio n, the Bureau of Meteorologys monitoring station on General Holmes Drive, Mascot and Sydney Airport Corporation Ltds monitoring station on Qantas Drive. Review and analysis of the datasets was undertaken in order to establish representativeness, quality and completeness. The Earlwood monitoring station is situated approximately 1.5km north west of Turrella and surrounded by similar topography to that in the vicinity of the Turrella stack. Discussions with EPA concluded that the 1995/96 data set be used for the numerical modelling as it was considered the most representative and recently

Design of the ventilation s y s t e m h a s b e e n undertaken by Hyder Consulting (Australia) Pty Ltd in order to establish the overall ventilation system configuration and determine airflows necessary to ensure compliance with the in-tunnel air quality goals under a range of operating scenarios. BHBBs sub contractor Alstom has carried out the detailed design of the mechanical and electrical components of the ventilation system. Having established the overall system configuration and airflows in order to meet the in-tunnel air quality goals, Hyder Consulting specialist air

available data set that achieved the appropriate degree of completeness and integrity. 7.2.3 Physical Dispersion Modelling

electrostatic precipitators prior to the opening of the tunnel to traffic. Documentation suitable for calling tenders has been prepared for the retrofitting of electrostatic precipitators including design, construction and maintenance. The scope of work includes all associated infrastructure and any adjustment required to maintain the integrity of the existing contractual obligations relating to operation of the ventilation system. This could include additional fan capacity to overcome pressure drops and to retain the operational flexibility and efficiency of the existing ventilation system.
q

MEL Consultants were commissioned by Air Noise Environment to undertake the physical dispersion modelling (wind tunnel testing) within the Monash University Boundary Layer Wind Tunnel on a 1:400 scale model (Mel, 1999). The modelling was undertaken for stack heights of both 25m and 35m under neutral, convective and stably stratified flow conditions in order to establish ground level concentrations of the stack emissions and to demonstrate the validity of the numerical modelling process. The hourly mean dispersion ratios derived from the physical modelling (for a range of efflux velocities and meteorological conditions) were correlated against the corresponding dispersion ratios under identical meteorological and efflux conditions from the numerical modelling. From the comparison it was found that the numerical modelling over predicted the mean ground level concentrations by a minimum of 2.0 times for both the 25m and 35m stacks. 7.3 Verification of the Design

Should the results of the monitoring show that the PM 10 contribution from the exhaust stack results in exceedance of the goals the RTA shall install electrostatic precipitators within 6 months of the direction by the Director General (or within other such time as agreed by the Director General). The RTA shall establish a protocol outlining procedures for deciding how an exceedance due to the stack will be determined. The protocol, which is to be made publicly available, shall be developed in consultation with the EPA and the Air Quality Community Consultative Committee (AQCCC).

In accordance with the project deed, the Contractor arranged independent design verification of the ventilation system design and the air dispersion modelling by Sinclair Knight Mertz. The RTA also engaged Tunnel Engineering Consultants of the Netherlands to independently verify the ventilation design. 7.4 Stack Height

A protocol has been prepared outlining a process for deciding if an exceedance is due to the stack. The protocol can be accessed on the RTA website at www.rta.nsw.gov.au An exceedance is not due to the stack if the exceedance is the result of an unusual event listed in the protocol including:

bushfires and prescribed burning natural causes (eg dust storms) fire incidents at residential or commercial premises traffic incidents involving a fire or spill illegal activities fireworks displays

Based on the outcomes of the numerical and physical dispersion modelling, the RTA submitted to DUAP the air quality assessment in April 2000 seeking approval for a proposed stack height of 25m at Turrella. On 22 August 2000 DUAP determined the height of the stack at a minimum of 35 metres. This determination included further approval conditions relating to air quality. These are summarised below together with the RTA response to the conditions:
q

The RTA shall prepare detailed plans and specifications for the construction of

The existing design has been under taken so as not to preclude retrofitting to the existing ventilation system.

The RTA shall establish a mechanism regarding the potential for complaints about air quality impacts resulting from the stack. If complaints are received from areas where there is reasonable potential for localised air quality impacts resulting from the stack and there is a reasonable correlation with stack emission levels, independent local monitoring of PM 10 shall be undertaken by RTA. If monitoring indicates local exceedance of goals, RTA shall undertake measures to meet the goals, mitigate concerns of the resident(s) or retrofit ESP. A mechanism has been prepared setting out a process to be followed if a complaint is received about local air quality impacts resulting from the stack.

A matrix has been developed setting operational concentrations and annual load limits for CO, NOx and PM10 in the stack. In addition, to address concerns raised by the community with respect to predicted NO 2 levels (albeit within the air quality goals) during hours 21, 22 and 23, minimum flow rates will be used under certain atmospheric conditions. The increased flow rates will apply during winds from the SE to SW sector when background ambient air quality concentrations of NO2 are greater than 50 g/m 3. Overall dispersion modelling undertaken by Air Noise Environment for Hyder Consulting predicts this will lower the maximum predicted receptor concentration (stack contribution only) of NO2 by 20 to 33% during these hours. As a result there will be lower receptor concentrations for approximately 90% of the 50 worst case predicted NO2 levels.
q

Subject to AQCCC agreement, RTA to fund establishment o f a c o m m u n i t y b a s e d monitoring station to monitor PM 10, NOx and CO. A community based monitoring station has been established as described in this paper.

The RTA shall also monitor PM 10, N Ox, temperature and volumetric flow rate in the stack. Equipment for monitoring PM 10, N O x , temperature and velocity of flow has been installed in the stack.

The RTA shall also assess and report on the impact of PM 10 stack emissions at monitoring locations in terms of meeting an annual average goal of 30 g/m3. The predicted annual average concentration for PM 10 is 16 and 19 g/m3 for the 1995/96 and 1998 data sets respectively. The predicted stack contribution is about 3%. This does not include adjustment for reductions of total emissions from i mproved traffic flow or reduction in local background concentrations from strategies implemented through the Air Quality Management Plan.

The RTA shall further investigate, in consultation with EPA, options for partial emission at tunnel portals to achieve energy cost savings as well as more widespread dispersal of emissions. Emissions shall not exceed air quality goals nor a CO eight-hour average goal of 9 ppm.
q

A study has shown that limited options exist for partial emission at the main portals (at the east portals more so then the west portal). Further studies will be undertaken when actual emission levels are recorded during operation of the tunnel.
q

The RTA, in consultation with EPA, is to develop a matrix of emission concentrations f o r P M 10 a n d N Ox and corresponding volumetric flow rates consistent with meeting air quality goals.

The Air Quality Management Plan (AQMP) is to include a detailed assessment of a buy-back or replacement scheme offered to all owners of solid fuel heaters in the local air shed. As part of the AQMP the RTA shall also undertake a detailed cost effectiveness comparison to assess the options for control of PM 10 and NOx. The AQMP shall also include a detailed education and communication strategy. See Section 11 of this paper.

7.5 Community Involvement

Under the above-mentioned 150 approval conditions two community liaison groups were formed for issues relating to the 4km tunnels. One was formed to have involvement specifically in issues relating to the tunnels (the Central Community Liaison Group CCLG) and the other community group to have an involvement in air quality issues, the Air Quality Community Consultative Committee (AQCCC). Nominated representatives of the local community groups and of each of the three local Councils have attended the meetings, which are held each month. Representation on the committees was by application with the selection criteria used to select representatives being;

developed in consultation with EPA and the AQCCC. It was required that the network was to be installed six months prior to operation of the tunnels. During the community meetings, concerns were raised as to applicability of the EPA Beaman Park air quality data to the Wolli Creek Valley. To address this concern RTA agreed to commence operation of the monitoring stations eighteen months earlier than required by the approval conditions. The initial station locations were selected based on:
q q q q

Availability of land Suitability of site to meet monitoring standards Site security Representativeness of site in terms of terrain and anticipated dispersion characteristics of the valley Input from the dispersion modelling of stack emissions

Local resident adjacent to the M5 East freeway Evidence of support of the community as a representative Availability to regularly attend meetings Willingness to communicate with the local community and present community views

There have been extensive presentations and discussions on all aspects of the tunnel ventilation system and air quality issues with the RTA, BHBB and Hyder Consulting responding to numerous issues raised by the community representatives. Extensive and detailed minutes of the meetings were prepared and provided to all committee members with specific issues retained on a rolling list until a response was provided. While agreement has not been reached on all issues, discussion has been extensive and vigorous. The draft air quality reports for both 25 and 35m stack heights were made available to the Air Quality Community Consultative Committee and the Central Community Liaison Group for comment. Responses by RTA, BHBB and Hyder Consulting to issues formally raised by the committees were in turn responded to as well as forwarded to DUAP for its consideration in determining the stack height. 8 AIR QUALITY MONITORING

Following discussion with the AQCCC, three initial sites were installed in June 2000. U1 the location at which the maximum stack PM 10 contribution was expected based on dispersion modelling. This location is to the north of the stack where the plume interacts with the higher ground on the ridge. T1 the location in the valley to the south east of the stack at which the maximum cumulative concentration of PM 10 occurs based on the dispersion modelling. T3 an AVOCs unit to the west of the stack to address the community perception of a build up of pollutants in the valley (possibly from Sydney Airport and other sources) and a sensitivity test for T1 to represent air quality conditions in the valley. Following DUAP approval of the 35m stack, an additional condition was introduced requiring installation of a Community Based Monitoring Station (CBMS). Following consultation a site on the ridge to the south west of the stack was agreed based on:

The project approval conditions require an ambient air quality monitoring network to be installed, including dedicated stations in the Turrella and Undercliffe areas. The location of the stations and the pollutants to be monitored are required to be

stations U1 and T1 could be augmented by data from the south west in terms of other potential

locations at which elevated concentration of other emissions such as NOx may occur

To provide independence from RTA, an auditor has been appointed to:

further air quality and meteorological data could be utilised to assist in determining if an exceedance occurred, is it due to the stack

confirm that the CBMS has been established and operated in accordance with relevant Australian Standards ensure the integrity of the reporting process by receiving, reviewing and reporting on monthly air quality monitoring data undertake 6 monthly audits on the operation of, and reporting from, the CBMS

RTA also agreed with the community to establish an additional monitoring station (X1) on the ridge to the north west of the stack. The monitoring network during tunnel operation and the emissions monitored is given in Table 1. Table 1 Emissions Monitored at Stations Station U1 T1 X1 T3 CBMS Location On ridge north of stack In valley to south west On ridge to north west In valley to west On ridge to south west Emissions Monitored PM 10, NOx, CO PM 10, N O x , C O , AVOCS PM 10, NOx, CO AVOCS PM 10, NOx, CO

The auditors report is forwarded concurrently to the AQCCC and RTA. The brief for the audit role, list of invited tenderers and selection of the successful tender was overseen by the AQCCC. 10 RESULTS OF BACKGROUND AIR QUALITY MONITORING Monitoring of ambient air quality commenced at the three initial stations in June 2000. Validated data from the EPA station at Beaman Park is available on a rolling basis, three months after the end of each month period. Data collected in the period June 2000 to August 2001 inclusive for PM 10 indicates:
q

Locations of the stations are given in Figure 3.

The readings recorded at T1 and U1 follow the same trends as at Beaman Park Generally the recorded readings at T1 and U1 are marginally lower than at Beaman Park In the fourteen month period from July 2000 to August 2001 inclusive the readings for PM 10 has exceeded the air quality goal of 50 g/m3 on one occasion, and exceeded 40 g/m3 on a total of only 6 days as shown in Table 2

Figure 3. Location of Air Quality Monitoring Stations 9 COMMUNITY BASED MONITORING STATION (CBMS)

These events were all associated with either bushfires or spring burns to reduce potential fuel for summer bushfires.

The CBMS has been installed and is being operated by Ecotech Pty Ltd, responsible for operation of all five stations in the air quality-monitoring network. RTA has provided funding for the station.

Table 2 PM 10 readings above 40 g/m3 July 2000 to August 2001 T1 g/m3 21/09/2000 05/10/2000 06/10/2000 07/10/2000 13/01/2001 16/01/2001 41.9 46.6 51.8 49.0 44.5 45.5 U1 g/m3 39.6 46.1 48.0 45.7 37.9 34.1

The RTA will contribute $2.5 million to the implementation of these strategies. This approach is in keeping with the generally accepted view that it is far more cost effective to address pollution at the source than to treat the effects. The AQMP includes an assessment of a buy-back or replacement scheme offered to all owners of solid fuel heaters in the local air shed. The AQMP includes an emissions inventory for the sub-region and has developed strategies to control and/or manage emission sources identified in the inventory. 12 CONCLUSION The M5 East ventilation system has been influenced by planning history, changing community values, community input and strict environmental requirements. An extensive process, with ongoing scrutiny and vigorous input from the community, has been undertaken for design of the ventilation system, taking into account all requirements and constraints on the project. Key factors expected for a successful project outcome during operation of the tunnel are:

Overall the readings to date indicate that:

the EPA Earlwood (Beaman Park) station in the Cooks River Valley is representative of the Wolli Creek Valley and, the background air quality and meteorological data from the EPAs Earlwood station in Beaman Park is appropriately representative for the design of the tunnel ventilation system and air dispersion modelling analysis carried out for the M5 East Motorway. When compared to the current data set acquired from these monitoring stations the 95/96 data set used for modelling purposes is marginally conservative. Rolling hourly real time monitoring data for NO2 and CO and daily data for PM 10 i s available on the RTA website at www.rta.nsw.gov.au . Any exceedance will be automatically flagged and reported through the Motorway Control Centre.

strict air quality goals (87ppm, 15 minute average at any location in tunnel, 50 g/m3 24 hour average for PM 10 at an external receptor). the small predicted increment in annual average levels for PM 10 of about 3%. conservative nature of the design verification of the design by international experts a higher stack (by 10m) than required to meet the strict air quality goals an expected downwards trend in vehicle emissions extensive air quality monitoring during operation with modification to the operation regimes of the ventilation system if required installation of Electrostatic Precipitators if an exceedance of the air quality goal for PM 10 is due to the M5 East ventilation stack

11 AIR QUALITY MANAGEMENT PLAN The Conditions of Approval require the RTA to work with the Department of Urban Affairs and Planning, EPA, Department of Transport and the Department of Health, to identify strategies and implement an Air Quality Management Plan (AQMP) for improving regional air quality. The aim is to identify key contributors to air quality in the area and formulate strategies to control these sources of pollution.

There are also expected gains in sub regional air quality from the Air Quality Management Plan. The Air Quality Management Plan also has the potential to be a model for use in other locations as part of strategies to address pollution issues impacting on urban ambient air quality. 13 ACKNOWLEDGEMENTS

The views expressed in this paper are the authors and do not necessarily represent the views of the organisation. The author acknowledges the team effort required in addressing air quality issues associated with the project, in particular personnel of BHBB, Hyder Consulting (Australia) and their sub contractors and RTA.

14 -

REFERENCES Hyder Consulting 2000. M5 East motorway driven tunnel services mainline tunnel air quality Mel Consultants Report 1999. Wind tunnel measurements of discharge from proposed Turrella ventilation stack