October, 2009
The primary focus of this submission is to propose an Alternative to the Metro 1 network that is considerably less expensive and offers more (but some differently distributed) benefits. An affordable Stage 1 alternative to the CBD Metro, that yields more early benefits, is also included in this proposal.
This Alternative requires an integrated rail network investment program that takes advantage of spare, but presently unusable, capacity within CityRail, and an exchange of traffic between CityRail and metro more in line with their appropriate roles. The capital cost savings arise from less new track being constructed, and not all of this being in tunnel, but with an adjustment for some additional costs associated with the exchange of roles. The additional benefits are in the form of better network outcomes. Complete operational separation from CityRail will be retained under this Alternative, and operating costs therefore should be similar.
Two separate papers, which have already been sent to the Minister of Transport and the rail corporations, accompany this submission. The first “Corporate Plans for Rail” levers off existing spare capacity to blueprint a multi-centred trunk public transport network within the Sydney region in support of the visionary metropolitan strategy. The second “Sydney Metro – Restoring Credibility” is a component of this blueprint that specifically addresses the proposed Alternative to Metro 1.
Metro was first mentioned by the Department of Planning as an intermediate public transport mode where deployment of heavy rail was not appropriate but conventional bus is, or would become, inadequate, and this view seems to have been adopted by Ron Christie for the metro proposals in his Long Term Strategic Plan for Rail. Under this scenario, metro would follow on from the Metropolitan Rail Expansion Program at some future time, and both its construction and operation would logically be completely separate from CityRail.
More recently, higher capacity Asian style metro has been proposed by the government, as a step change from CityRail, to enable metro to be part of the present rail network expansion plans, rather than just a future complement. These different views explain how Les Wielinga and Ron Christie could debate metro from opposite corners on Stateline recently and be both right. The new approach is assumed by this submission as outlined next, however it is also apparent that the previously logical complete separation of metro construction would now be detrimental. A further logical consequence of such a step change from CityRail to Asian style metro is that the key need for an intermediate mode for some corridors still needs to be addressed.
CityRail performs poorly on its inner line services. These are both slow and infrequent due to the universal use of double deck rolling stock and, in many cases, from the pressure to accommodate more outer suburban commuter services to the Sydney CBD over sections of track shared with these services. When the MREP, including a new CBD heavy rail link, was announced in 2004, there would have been an opportunity to partially reconfigure CityRail by introducing metro style trains on the four inner lines serving the City Circle, and, with some junction enhancements, migrating outer line services from the City Circle onto the new link as a separate sector. While an improvement on having all double deckers, it would have left the northside with excess heavy rail and no metro style capacity. This didn’t happen, and CityRail has reportedly looked at single deck trains again, but with a compromise design maximising seating capacity, to address capacity problems on the Liverpool and Harbour Bridge lines. Again, nothing has happened, presumably because the present sectorisation by route, rather than also by distance, would only make it possible to partially convert any one sector to single deck operation.
Metro has the potential to introduce the additional sector needed to reconfigure, by distance and also to expand, Sydney’s rail network, so a step change to metro should be good news. Being a completely separate sector also allows modern technology and a new operator to be introduced, provided users are presented with an integrated fare structure and convenient interchange. Introducing metro should also be compatible with the metropolitan strategy, with its ethos of “jobs closer to home”, as this implies less emphasis on long line commuting to the CBD and more emphasis on frequent shorter distance services (and also the emerging cross regional trunk role that is a separate issue) in the future. However, overcrowding on existing outer line services still needs to be addressed by providing some additional capacity into the CBD in the short term, while putting a limit on the amount of further longer term growth.
While a step change to metro is good in principle, the present CBD Metro proposal looks odd. Serving Rozelle is hardly a current transport priority and there is no present zoning that requires such a large boost to capacity. Having more outer line users change trains at Central seems an imposition; especially since Bradfield’s farsighted city railway plans were designed to avoid repeating this historic feature of the rail networks in other cities. Perhaps this would be OK if it was the first stage of something less odd, even though limited funding and energy availability may mean being stuck with the first stage alone for some time.
Unfortunately, the full proposal for Metro 1 looks even more odd. Journey times on West Metro will be faster than CityRail and metro is planned for much, and possibly all, of the route from the CBD to the northwest that is essentially an outer line commuter service. Further, CityRail would shed very little of its inner line patronage and only radial routes to the CBD are being proposed. All this is far from the evoked relationship between main line and metro from London and Paris, where main line services are faster with fewer stops and metro provides two-dimensional local coverage over a mesh of routes.
An integrated rail network investment program would enable the above oddities to be addressed, along with the realisation of additional benefits and cost savings as already mentioned, by taking advantage of spare capacity within CityRail and an exchange of traffic between CityRail and metro. This alternative approach is described in the accompanying paper “Sydney Metro – Restoring Credibility” and is summarised in Annex 1 of this submission as a list of features of, and benefits from, each stage of the proposed Alternative, accompanied by “before” and “after” track diagrams.
Using the existing Local Line between Strathfield and Newtown would save about 9 km, or about $3 billion, of new underground construction based on a pro-rata cost estimate with West Metro. This saving would be reduced by the need for expenditure on metro conversion and CityRail junction works, but should still exceed $2 billion.
Although the length of new track from Martin Place to Gordon under the proposed Alternative is similar to the track length to Epping under Metro 1, six km less would be in tunnel and there would be around five less underground stations. Again, the saving would be reduced by the need for expenditure on metro conversion, but a net saving approaching $1 billion seems likely, bring the total saving to about $3 billion.
A cost estimate for the 10 km of new construction under stage 1 of the proposed Alternative can be built up from the cost of the 7 km CBD Metro, an additional 3 km of new underground metro with two underground stations, the need for expenditure on metro conversion over the other 9 km of surface track, and CityRail junction works. While higher than for the CBD Metro, this $6.5-7.5 billion estimate is budget-affordable, and should be preferable considering the earlier delivery of rail network benefits otherwise dependent on completing the currently unfunded West Metro.
There will also be a future saving at the time a Warringah metro is being considered, as this project would be able to commence from the Mount Street branching point, rather than the CBD. Strictly, this saving should be reduced by the net cost of the matching Airport Line conversion (and out to Riverwood or Revesby) after operating cost savings are considered.
While the proposed Alternative leads to a more economic outcome, there are transition arrangements, due to the integrated investment, that need to be considered. This is because making changes to existing infrastructure must inevitably lead to service interruptions being needed, meaning that implementation will be less straight forward than for an independently established Metro 1. Annex 2 canvasses a range of transition possibilities, but does not identify a preferred solution.
The following observations illustrate the impact of taking an integrated network, rather than an individual mode, view of transport planning.
The EA claims that the CBD Metro and the rest of Metro 1 will create opportunities to reconfigure the CityRail network by potentially unlocking spare capacity. However, there are no definitive plans to do this; only a note that further analysis with CityRail is being undertaken. It also seems that it will be mostly CityRail having to adapt to Sydney Metro, rather than a mutual optimisation, as there is no specific mention of adhering to the role split between longer distance commuter travel and higher frequency shorter trips. The oddities mentioned previously also give this impression by indicating that, at a network level, this role split is only loosely being followed.
The proposed Alternative presented in this submission actually includes a reconfiguration of CityRail to unlock spare capacity, rather than this being an additional network cost under Metro 1. What’s more, much of this unlocking would now occur in the proposed first stage. The proposed Alternative also more closely, and more credibly, follows the longer/shorter role split mentioned above. Effectively, the built-in CityRail reconfiguration is a further economic benefit of the proposed Alternative when compared with Metro 1, and confirms the need for integrated rail network planning and investment. There is also some doubt, evidenced by CityRail’s study of single deck trains to increase capacity, and the scale of some of the infrastructure work envisaged for the proposed Alternative, that CityRail would be able to easily reconfigure on its own anyway.
The EA makes much of the favourable impact of CBD Metro on the Victoria Road corridor and its bus services. However, an intermediate capacity mode would be adequate for this corridor, if considered in isolation, and would also extend further along it. Accordingly, this would seem to be a better match for the potential demand than the much more expensive, higher capacity and fully underground Metro 1. The picture is more complicated when the metro extension to Epping is considered, as this faster route would draw additional patronage, partly from CityRail, but at the same time away from the route through Chatswood to thereby weaken support for the metropolitan strategy and its global arc. The proposed Alternative, on the other hand, would more fully support the metropolitan strategy and provide the capacity for a Warringah metro to be constructed from North Sydney.
With respect to an intermediate capacity mode to be deployed on the Victoria Road (and other) corridor, there are many worldwide examples. These include trunk bus in many forms, light rail (often branded as metro) and automated light metro such as the DLR in London and the new Ansaldo STS system in Copenhagen. Accordingly, it is not technically valid for Sydney Metro to claim (in the Eco Transit response) that there’s only a choice between metro or buses for the Victoria Road corridor.
Sydney is known to have much more road space per capita than Asian cities, and most European cities, and accordingly underground construction is not the only feasible way to relieve congestion. The alternative is making better use of existing road space by introducing time competitive trunk public transport on a dedicated and segregated right of way, as has already been achieved by many more densely populated European cities through less car use, despite similar per capita levels of car ownership. Grade separation would only be used where needed at intersections, stations and crowded areas, and where the surface alignment was unsuitable. Further discussion on intermediate mode deployment is provided under “Light Metro” in Section 4.4 and Annex D of Corporate Plans for Rail. The NSW Treasurer’s ambit claim for a “tunnelling premium” as part of the GST dispersion looks dubious in the context of being able to make more efficient use of existing road space in this way.
It is also important to recognise that further transport expenditure on the Victoria Road corridor would not necessarily be as high a priority without the metro plans, and that the first stage of the proposed Alternative will allow Burwood, Ashfield and Petersham to be strengthened as modal interchanges to reduce the number of buses along George Street
Sydney’s low population density suggests a limited applicability for new high capacity rail corridors. There are also looming climate, energy and budget imperatives that will impact on both the need for more public transport, and the ability to respond to this need in a reasonable time frame. Accordingly, it is reasonable to expect that these issues will constrain the amount of new fully underground metro that can be constructed, and that the real choice for some radial and cross suburban corridors is more likely to be an intermediate capacity mode or nothing.
Of the two observations, the first clearly supports the integrated investment approach used for the proposed Alternative, while the support is more marginal for the second. Additionally, an integrated investment program would presumably be contingent on the CityRail component being appropriate, as it has a reputation for slow and highly subsidised services. The slowness relates to using double deck trains on short hop services, the mixing of different stopping patterns, infrastructure limitations, the traction performance of the pre-Tangara air-conditioned trains, and complexity generally. These speed issues in turn limit the utility of the service and the fares that can be charged for outer line commuting. There is also a (presumably consequent) history of funding availability counting against CityRail, as illustrated by the recent unbalanced emphasis on tollway building ahead of much promised public transport.
More positively, the new double deck trains that are being supplied to fully air-condition the CityRail fleet are well suited to the outer line commuting task, and Sydney Metro acknowledges the need for this role to continue. Additionally, the focus of the integrated investment under the proposed Alternative is to improve outer line commuting, divert slow inner line service patronage to metro, and improve infrastructure. Further, and despite past lapses, Treasury procedures are supposed to require project evaluation and selection to be done ahead of funding availability considerations.
The metro concept has evolved from being an intermediate capacity mode that was complementary to the Metropolitan Rail Expansion Program to one also directly addressing the CBD capacity constraints on the present heavy rail network. As such, metro has now been made an integral part of the present strategic need to expand this rail network. The current Metro 1 plan, and its CBD Metro first stage, exhibits aspects of each of these evolutionary stages.
With hindsight, it is therefore not particularly surprising that a more economic Alternative, as proposed in this submission, would move metro closer to being part of the original MREP configuration that was designed to support the global arc within the metropolitan strategy. Further, the introduction of metro this way brings more benefits, and a better structural separation of long haul and short haul services, than was envisaged under the MREP. However, this now leaves the original issue of an intermediate capacity mode still to be addressed. With climate, energy and budget imperatives looming large, and the urgent need for more public transport, it seems unrealistic to believe that fully underground metro can be the universal solution to Sydney’s future non-bus transport needs. Accordingly, some serious investigation of intermediate modes appears to be warranted.
This Annex lists the features and benefits of each stage of the proposed alternative to Metro 1. The list should be read in conjunction with the “before” and “after” track configuration diagrams that follow.
| Capacity (trains per hour per direction) available as follows: | |
|---|---|
| Total for Suburban and Main Lines | 40 |
| North and west in 2009 timetable | 26 |
| Spare for the above | 02 |
| South in 2009 timetable | 08 |
| Sub Total | 36 |
| Additional capacity available | 04-06 |
While the proposed alternative leads to a more economic outcome, there are transition arrangements that need to be considered due to the integrated investment. This is because changes to existing infrastructure must inevitably lead to service interruptions being needed. Key issues are retrofitting metro power and control to sections of CityRail track and the intricate junction works in the vicinity of Newtown.
A range of transition possibilities for the Silverwater to Martin Place first stage is presented below, but a preferred solution is not identified. A common factor is that all assume the two Bankstown and two Liverpool via Regents Park services per hour will use the Lidcombe turnback and only the six via Granville services from the south will proceed further east, in response to the Local Line being closed to CityRail.
Firstly, reactivation of Illawarra Junction for a temporary period could be considered, despite a potential loss of network reliability. If so, the six via Granville services per hour from the south would use the Suburban Line, while more services from the west would use the Main Line and possibly some also end at Sydney Terminal. Some Suburban Line services would stop at Burwood, Croydon and Ashfield, and also Newtown after a short interruption to slew the Suburban Line over to the existing platforms in preparation for the junction work. This would leave only Summer Hill to Stanmore to be served by buses to the CBD, and Macdonaldtown to be served by a local bus shuttle to Redfern, during the construction and conversion period.
Secondly, the present sectorised operation could be retained by reconfiguring Homebush, so that the turnback is reversed to terminate the six via Granville services from the south. Additional trains could be run between Olympic Park and Sydney Terminal to serve those changing at Homebush, although some passengers may prefer to change at Lidcombe onto already crowded services from the west. Suburban Line stopping patterns and replacement bus services would be as the previous possibility. A supplementary City Circle shuttle service may also be necessary.
Thirdly, the Local Line interruption period for the Summer Hill to Stanmore section could be much reduced with a two-phase implementation of the first stage. The key issue is that the Local Line service needs to cease before the Suburban Line can be slewed at Newtown and the junction work commenced, but the proposed final resectorisation under the alternative can only occur after the junction is complete, leading to an extended interruption period. The first phase would be to operate metro between Martin Place and one Local Line platform at Strathfield, with the six via Granville services terminating at the other, for cross platform interchange. The second phase, from Strathfield to Silverwater, would have been constructed, but would be commissioned only after the Newtown junction work was completed. Although Macdonaldtown would be closed during and after the first phase, the nearby metro station at Darlington would then be operating. A further complication with this possibility arises because the Silverwater depot would only be permanently connected to the operating metro after the second phase. Accordingly, interim arrangements would need to be made to store metro trains in the first stage tunnels and transfer them manually from Strathfield into the tunnel portal for getting to the depot only when needed for maintenance. Probably only one of the two tunnel portals would be accessible for this transfer, while the Strathfield to Homebush section of the Local Line remains under CityRail use, given the limited space available.
Extended periods of replacement bus services will also be needed for intermediate stations during the North Shore metro conversion, and this can be implemented in three steps. These are to St Leonards, to Chatswood and to Gordon. Firstly, the new CityRail running lines would need to be constructed for each step and cutover into traffic. Following the cutover, the intermediate station(s) would be isolated until metro service was established, as CityRail would then only be serving the key stations. Accordingly, replacement bus services between pairs of key stations to serve the intermediate stations would be needed in turn for each step of the conversion.
October, 2009