Wellington Scoop

An affordable option: rail and road in one tunnel


by Glen Smith
Planning is being undertaken for a possible light rail spine linking Wellington Railway Station with Wellington Airport. The evidence supporting this is compelling. And there is an option for overcoming one of the key obstacles – an affordable and minimally destructive route across Mt Victoria.

By the time any construction begins, it will be almost a century since the transport capacity across Mt Victoria was last increased, with construction of the Mt Victoria Tunnel in 1932. With ongoing growth this is now inadequate and further capacity is necessary. This increase can’t be supplied by overland routes – new tunnel space is required.

The Spine Study recommended adding a second two-lane Mt Victoria Tunnel for road traffic at a cost of around $370million. It also explored the option of a dual-track rail tunnel to the south of the existing Mt Victoria Tunnel at a estimated cost of $380million, but noted that additional land would likely be required to the south of Patterson Street and, since this area is occupied by several schools, concluded that this would have ‘considerable effects throughout this area’.

The study also briefly looked at combining rail and road in a single tunnel but in a report that listed reasons why it was felt this wasn’t possible, stated that “we are not aware of any LRT/road tunnels that have been built or operated elsewhere” and that “rail and road traffic is always located in separate tunnel bores”.

The source and qualifications of the person making this statement weren’t identified. A quick internet search shows the statement is false, with a number of examples around the world of multimodal tunnels with different modes travelling in physically separated compartments of a common tunnel bore or structure.

As part of this search I came across the Fehmarnbelt Tunnel, a stunning 17.6 km combined road and rail tunnel planned by the Ramboll company up to 40m deep under the sea between Germany and Denmark.


In 2013 I wrote to Alun Thomas, then head of tunnelling at Ramboll (but since headhunted by Minova to direct tunnelling for Europe, the Middle East and Africa), enquiring if he felt a dual road/rail tunnel was an option here in Wellington.

Given that he must deal with contracts worth billions, I didn’t expect him to reply. But he did and as a general response stated that ..“in short there is no reason why a tunnel can’t have multiple uses..”. He noted that “..we have just …proposed a road / metro tunnel here in Copenhagen to cross the harbour.”

I had initially proposed a modification of the NZTA’s tunnel profile with a single rail corridor replacing the cycle/ pedestrian lane and cyclists /pedestrians either hitching a free ride or travelling via a separate parallel tunnel.


With accurate timetabling this would involve no train delay. With random train arrivals and tunnel transit time of one to one and a half minutes (depending on speed/ acceleration/ deceleration) the average wait time would be 10-15 seconds. For safety I suggested cross tunnels to the existing Mt Victoria Tunnel as escape routes.

But Alun suggested a stacked design with LRT on one level and traffic on the other. He suggested LRT on the top, but gradient considerations would make LRT on the lower level easier.


He noted that the tunnel would be very short by international standards. He said safety considerations could be overcome and felt it was likely no cross passages would be required since the tunnel was so short. He cited a number of tunnels around the world with similar stacked design where safety and other issues had been overcome.


He concluded that this was a viable option for Wellington, recommended that this be investigated further, offered to undertake a feasibility study and suggested that a meeting could be arranged with one of his colleagues who was due to visit New Zealand at the time.

To explore this option further I wrote to the NZTA and the Regional Council. Their response could generously be described as… unenthusiastic. The Planning Advisor (RoNS) for the Wellington Regional Branch of the NZTA stated:

“..we have not carried out any investigations, designs or costings.” for such a tunnel. The reason given was that “what you are interested in is a type of tunnel that the Transport Agency is not able to deliver as we can only designate for the purpose of road, construction, operation and maintenance (including pedestrian and cycle activities). Public transport needs are provided for by Greater Wellington Regional Council.”

This statement can only be seen as a stunning indictment of our planning processes and confirmation that the NZTA is a ‘transport’ agency in name only.

I received no reply from the GWRC initially, so sent an official information request asking for details of any “possible designs, engineering and technical reports, and costings for any dual or multipurpose tunnels through Mt Victoria that the Wellington Regional Council has considered as part of their planning process…”.

After a delay, then chairperson Fran Wilde replied that only a ” shared tunnel option (Light Rail Transit (LRT) with other vehicles) and a separate tunnel option were considered” and that the “shared tunnel option was based on LRT vehicles using the same tunnel bore without any form of separation.” The reason given for not considering physically separated corridors in a common tunnel bore was that “verbal advice was received that enlarging the tunnel bore to provide additional space for separation measures would likely have similar costs to that of a separate tunnel bore — resulting in a similar benefit cost ratio.” The identity and qualifications of the person giving the verbal advice was not given and on further official enquiry could not be supplied.

I asked Alun whether he felt this conclusion was valid. He answered that “the reply citing verbal advice sounds weak. For comparison, please find attached a paper describing costs in Figure 5.”


And he continued

“…if we are talking about 2 x 10 m diameter tunnels, one for road and one for a twin track LRT, this would probably be more than the cost of a single 14 m tunnel containing both the road and the LRT tracks. (e.g. taking the average cost line, 2 x 22 = 44 vs 1 x 39). Changing from a 10 m to 12 m diameter tunnel to accommodate a separation wall certainly wouldn’t cost as much as building a new tunnel.”

I suggested to him that “even if the cost were similar, a single dual tunnel would be far less destructive in the approach on either side.” He replied “I think that you are absolutely right about the additional impact of multiple tunnels on the densely populated approach areas. This is a significant advantage of a single bore.”

I noted that pedestrian and cycle corridors would also be required and asked whether all modes, including cycling and pedestrian, could be included in separate compartments of a common tunnel bore. He replied “yes – this combined option should be economic and simple”. A quick look at the approach areas shows that accommodating rail, road, cycling and pedestrian corridors should be straightforward:



It’s now over five years since this option was first presented, and that period has seen the establishment of the Ngauranga To Airport Governance Group and the Let’s Gets Wellington Moving Governance Group, partnerships between the WCC, GWRC and NZTA tasked with exploring options for transport across the city. We might expect such groups would be able to undertake the exploration of options in a more thorough and objective manner. Sadly there is no evidence that this is the case.

The four ‘scenarios’ developed by the LGWM group were essentially predetermined stages in a minimally altered rehash of the deeply flawed ‘Spine Study,’ with none of the profound basic design flaws addressed. This included no exploration of options for removing the transfer penalty at the Station to produce a seamless network, no consideration of adding a Quays-based rail corridor to avoid bus displacement along the Golden Mile, no Option X design options for the Basin Reserve despite its logic, and the only tunnel option considered being “an extra Mt Victoria tunnel including walking and cycling access”.

The only light rail route offered in public survey questionnaires was a route via Newtown, despite Newtown being adequately serviced by buses and this route being longer, slower, technically difficult, almost certainly more expensive, forcing all commuters from the East to unnecessarily go via congested Newtown, forcing unnecessary transfers on a significant number of bus commuters and displacing car capacity and parking which will attract likely ferocious opposition from shop owners and residents. The result will be a protracted consent process and, if it goes ahead at all, lower PT uptake.

It is clear that rather than undertaking a thorough open public consultation process in which all options are explored and presented, our planners are working on a predetermined agenda. And this agenda involves ignoring the recommendation of a leading world tunnelling expert on the most direct, fastest, cheapest and least destructive way of achieving a high quality dual rail corridor across Mt Victoria.


  1. Guy M, 8. February 2019, 10:03

    Glen – interesting post. I can’t help but think that a combined tunnel would be unlikely, due primarily to the size and the length. Auckland’s Waterview tunnel was unusual in that it was large – super large – the 4th largest (diameter) TBM in the world dug the hole, for 3 lanes of traffic. Volcanic soil too, so some hard, but easily fractured rocks, but the main thing is that it was long and continuous and they could have a good sized yard at the beginning and end for a TBM (tunnel boring machine).

    We have a very different situation in Wellington. We have some really short holes to dig, in the middle of an already crowded city, so it is difficult to find staging room for any TBM equipment. Plus, we certainly don’t need to have the train line going down the same route as the car route. With some good sense and good planning, the two routes can and will be completely independent.

    No real reason therefore to have a combined rail/road route. We’re not on a bridge, nor in an undersea tunnel, so the routes can go wherever they please.

  2. Keith Flinders, 8. February 2019, 11:37

    A different approach Glen , but it doesn’t resolve the need for grade separation to counter Basin Reserve traffic congestion. A further tunnel under or alongside the Basin Reserve presents engineering challenges due to the high water table, a flyover in this location is not acceptable to a large number of Wellington residents.

    Ideally a second road tunnel should go further south, starting at the top of Wellington Road and finishing the southern end of Wallace Street. New tunnel to carry two lanes of east – west traffic, existing tunnel two lanes of west – east traffic. This would reduce the Basin Reserve congestion issue substantially.

    I prepared such a scheme when standing for the GWRC in 2016. Not a cheap option, but then value never comes cheap, however it would serve the city for many decades as the existing Mt. Victoria tunnel has done. Getting a light rail system up and running, then convincing motorists to use it instead of their cars will be less expensive and better for the environment.

  3. Reg Varney, 8. February 2019, 11:41

    A trackless tram and one tunnel will suffice.

  4. Glen Smith, 8. February 2019, 13:24

    Guy. Alun had not suggested a TBM but roadheader or drill and blast (presumably the methods the NZTA were planning to use). Large tunnel bores are now quite common world wide and, as Alun says, cheaper and less destructive. What logical reason do you give for road and rail not travelling together?? They are servicing the same population distribution but by different modes. (The main trunk line follows SHI for large distances through the North Island – perhaps we need to move the main trunk line?) The routes can’t go wherever they please – rail has to go through Mt Victoria at some point which involves large cost for the tunnel and destruction on the approaches.
    Keith. I hadn’t suggested a tunnel under the Basin or a flyover (which I appeared before the Basin Board vigorously opposing). Full grade separation at the Basin can be achieved at the north west corner of the Basin by means of a short trench at the eastern end of Buckle Street as part of an Option X layout which is almost all at existing ground level (see previous design in my Basin article). The Tunnel I outline here would be at current Mt Victoria Tunnel height (or slightly lower depending on rail gradients).
    How much would a Tunnel from Wellington Road to Southern Wallace street cost? How many key destinations does this fail to service? And how would you get rail from the city to the southern end of Wallace Street? (insufficient width on Wallace St for dedicated rail and road even if you remove all residential parking).

  5. TrevorH, 8. February 2019, 13:30

    Has anyone seen the “compelling evidence” for light rail in Wellington? Maybe my copy is still in the mail? [You missed all these articles? Undeniable: the case for light rail.]

  6. Keith Flinders, 8. February 2019, 16:07

    Glen: What I wrote in 2016 is still at http://briefs.flinders.nz/ A series of observations including the Basin Reserve traffic issue and maps. I never considered mixing rail or light rail in the same tunnel/s as your concept does.

    I don’t see heavy rail as an option through or adjacent to the CBD, but others do. I do see light rail (the best option at present but this might change) to Miramar/Airport via Adelaide Road and a new tunnel at Mt. Albert, but not in my lifetime.

    Yes on street parking along Wallace Street would need to be sacrificed, not that I expect residents of that area would be too thrilled at the prospect. In time much of the area Wallace to Adelaide is going to be redeveloped into higher density housing and will require mass public transport services. Much of Newtown will be redeveloped over coming decades.

    With NZTA and the GWRC involved, we will only get short term planning results, not the longer term vision of those who planned the Mt. Victoria tunnels and the tram systems of the early 1900s. Karo Drive being a prime example of more recent short term planning.

  7. TrevorH, 8. February 2019, 16:38

    Dear Editor: I have seen many of the articles to which you refer but I have yet to see a sound business case in support of light rail from the railway station to the airport. Lots of very detailed track plans though…

  8. Roy Kutel, 8. February 2019, 18:22

    TrevorH – GWRC’s Business Case estimated Light Rail would return 5 cents in the dollar. If you live in Wellington City, your rates would go up by $173 a year to fund Light Rail which is on page 69 of the following report.

  9. Trevor H, 8. February 2019, 19:06

    Thanks Roy for these important references. I am concerned however that Mr Twyford wouldn’t recognize a business case if it bit him on the bottom.

  10. John Rankin, 8. February 2019, 21:03

    No, @RoyKutel, the business case estimated that a particular light rail project would return 5 cents in the dollar. Given that the proposal broke every best practice guideline for light rail projects, it’s surprising the business case wasn’t worse. You can’t conclude that other light rail projects will have the same rate of return.

    Different road projects have different rates of return and we only build the ones that have a high enough rate of return. For example, NZTA is not building an interchange at Peka Peka because the business case doesn’t stack up. By your logic, NZTA would never build any interchanges, because the Peka Peka interchange business case is not good enough. This is clearly nonsense.

    The lesson we can learn is to identify what makes light rail projects succeed or fail, then consciously set out to design a project with the maximum chance of succeeding.

  11. Roy Kutel, 8. February 2019, 21:48

    @JR – TrevorH asked for a Business Case so I referred him to GWRC’s $1 million dollar Business Case undertaken by AECOM. The peer reviewer agreed with the process and the results.
    Regarding road projects, the Labour/Green Party (the Govt before John Key’s National one) removed the requirement for transport projects to have a BCR > 1 to get funding approval which allowed the Roads of National Significance to happen. Criteria such as ‘strategic fit’ and ‘effectiveness’ became important.

  12. Glen Smith, 9. February 2019, 9:28

    Trevor and Roy. I agree that no cost benefit analysis that demonstrates the benefits of rail has been produced. The only ones I have seen are the farcical ones that Roy references.
    Any proper analysis would include the costs associated with the car use that commuters are forced to undertake due to the lack of a high quality attractive PT alternative. These include congestion (likely hundreds of millions within the next couple of decades likely rising to billions over this century – see previous article threads), cost of road building to accommodate the extra cars (billions to tens of billions depending on the timeframe), use of valuable city land that car users feel they have a god-given right to occupy for free (30% of city land – work out the cost of that), death and maiming in accidents from the car trips commuters are forced to take ($4.16 billion in 2016 – nearly $1000 per person or $4000 for an average family of 4), pollution from gas belching vehicles (over 1100 premature deaths per year as cost of over $4.28 billion), policing of roads, and almost certainly the largest but as yet unquantified cost – the long term cost of climate-change from carbon the vehicles produce (how much will it cost to move the CBD to Tawa once sea level rise sets in?). None of these are included in the farcical analysis you reference.
    Fortunately I am sure that, unlike commentators who lack the necessary intellectual prowess, Phil Twyfods is aware of these legitimate costs and will factor them into decisions. Anyway, enough of the narrow blinkered bleating from naysayers and back to the practicalities of producing the high-quality across town rail corridor that 63% of Wellington survey respondents say they want.

  13. Guy M, 9. February 2019, 9:38

    Glen – thanks, I agree, a road header would make much more sense. Re your comment “What logical reason do you give for road and rail not travelling together??” – seems to me that a rail line, of any sort, would need to go from the railway station, uninterrupted, to the CBD, Te Aro, Hospital, Airport, with station stops at each of those places. But a road tunnel would want to be part of a network which links up to incoming and off ramping side roads all through the trip, and could well take a completely different route.

    There just doesn’t seem to me to be any logic to force them to travel together for any part of the journey, and especially at the most congested point on the entire trip, i.e. the Basin. Keep them on entirely separate routes, build smaller tunnels where they are needed. Your comment about SH1 travelling near the main trunk line illustrates my point – while they may be nearby, they are not directly on top of each other, and have their own separate routes.

    There is certainly a lot of logic to say that a Light Rail route could / should go up Taranaki St, which is certainly wide enough, and gets rid of the issue at the Basin. Using a solution akin to Option X at the Basin for the cars means that the traffic issues there can get resolved a lot easier. A second Mt Vic tunnel for cars should be dug anyway, not just for traffic reasons, but also because the existing one is below standard for pedestrians and cyclists, due to fumes and idiots tooting. Retaining the existing Tram tunnel for the existing retained buses takes care of that issue (i.e. separate buses from cars) and keep the new proposed LR on its own separate route.

    Keep up the good work.

  14. Glen Smith, 9. February 2019, 9:43

    Keith. An interesting option which I hadn’t seen – the more exploration of options the better. I could see a few hurdles. The length of tunnel/ underpass would be almost 1300m including going under or ? destroying the historic building around John Street and coping with the services (likely a lot), the gradient required to get this underpass back up to Wallace St level (John St already has quite a gradient) and the difficulty in removing all the parking on Wallace Street. Your plan uses Wallace St (already a busy road with shops/cafes/ side roads/ pedestrian crossings/ university entrances) to traffic from the east to the city – do you think this is a sensible option compared to a dedicated route along a widened Ruahine St. I couldn’t see you had incorporated rail; was this in your plan?

  15. Jonny Utzone, 9. February 2019, 9:47

    Guy M – The Wellington Motorway goes directly above the railway on a flyover before Kaiwharawhara.

  16. Glen Smith, 9. February 2019, 10:04

    Guy. Interesting points. There is no particular reason why road and rail have to follow the same route but if the best route for rail happens to be the same as for road (which I believe it is) then why would you chose a different rail route just because the road happens to be there? If at the same time it is also cheaper, faster, more direct and less destructive then putting the rail somewhere else doesn’t make logical sense.
    You indicate that rail has to pick up every destination, ignoring the fact that we have a comprehensive bus network servicing these destinations already and which rail won’t be able to replicate (and we shouldn’t try). The aim of rail, in my view, is to provide a high quality corridor for across town commuters who are put off by the Station transfer and across town slow bus trip down the Golden Mile. A key aim should be to remove both of these barriers. This should pick up as many destinations as practicalities allow (and I believe we can pick up most) without compromising the high quality of the corridor- any remaining ones can be serviced by bus lines. Taranaki St would be nice but faces the barrier of getting east from Taranaki St would would require an expensive Tunnel. Picking up the hospital would be nice but the barriers to a high quality Newtown rail corridor are huge and would put off a large number of commuters from the east- Newtown and the hospital can be serviced by bus.

  17. Glen Smith, 9. February 2019, 10:08

    Keith. I wasnt planning ‘heavy’ rail (depending on your definition of this) but a ‘lighter’ engineered corridor (for ‘lighter’ units tracksharing on the main network) although one that is still very high quality (dedicated and almost fully segregated with all intersections fully light controlled with full rail priority). I think this is achievable

  18. TrevorH, 9. February 2019, 14:05

    If rate and/or taxpayers are to invest in light rail in Wellington then a business plan set over a specific time-frame needs to be produced. From the many posts on this site there are a great many assumptions made about the benefits of light rail but very little evidence other than attempts to draw parallels with other cities overseas which to begin with almost always have much higher population densities along their tramway routes. So perhaps it’s time to take a break from track-planning and get out those calculators or abacuses as the case may be.

  19. Guy M, 10. February 2019, 9:01

    Sigh… I don’t want to get into a slanging match, but I have worked on a large underground rail project, so have some experience in what I am saying: there is no reason for Wellington to combine rail and road in a tunnel project, and it would make things so much harder than having two separate structures. I’ll leave it there.

    Trevor H – yes, Wellington’s population is at the smaller end of cities with LR, but it is not off the scale. Still very doable.

  20. Keith Flinders, 10. February 2019, 11:20

    Glen. When I saw Hataitai Station on your concept drawing, it gave me the impression that you were thinking of heavy rail. I expect you were thinking instead of the train-tram concept as used in Karlsruhe, Germany. That being lighter passenger rail vehicles which run on standard rails and light rails. Taking any land off the town belt alongside Ruahine Street will be met with years/decades of opposition.

    I agree that the slope up John Street to Wallace Street in my concept plan would be quite steep, although not an issue for cars. Most historic buildings in the area are timber framed and can be relocated. From Adelaide Road to the start of Wallace Street the rise is 19 metres. If an underpass was created under Adelaide it would add another 4 or so metres, depending upon the current placement of sewer and storm water services under Adelaide.

    If a Tunnel Boring Machine was employed, then rather than stopping at Adelaide keep going to Taranaki Street, or further west. Wallace Street then remains as is. All able to be done engineering wise, cost obviously a factor, but we need to be planning for beyond the next ten years to the next 80 or so years. By then the eastern suburbs population might well have doubled, or more, with higher density housing.

    NZTA are unlikely to agree with me, or indeed any of us who use lateral thinking. As was amply illustrated with their Basin Reserve flyover, no other option could be entertained. They lack the long term vision this city needs.

    Unlike FIT I see light rail through the CBD replacing many of the buses that currently run Lambton – Willis – Manners. Think of light rail as a long bus. Cars and other vehicles banned from Lambton Quay in peak hours, being controlled by automatic congestion charging and as used very effectively in central London. Use the congested city routes in peak hours and pay $50 for the privilege. Could be implemented now at far less cost than changing lane markings, etc.. We don’t have room for more lanes/roads and must make better use of what we have by rationing it.

    I see the light rail route passing through Newtown to Kilbirnie and to Miramar/airport where most users are close to it, with priority traffic light control and removal of some on street car parks. I don’t subscribe to the notion that we need a fast rail/ light rail service to the airport. Faster, yes. An extra 5 or 10 minutes penalty for a limited number of passengers requiring a speedier trip to/from the CBD would require a route unable to maximise potential capacity.

    Finally I will throw this idea into the mix. Light rail, or any other form of mass public transit, to use Tory and Tasman Streets going south only. Both streets become one way, sharing with other traffic, and parking limited to one side of the streets only.

  21. Kerry, 10. February 2019, 15:52

    Trevor & Roy – The business case in the Spine Study was useless but — throughout the study — the real case for light rail was hidden in plain sight.
    The Spine Study put up a case for a second bus route through the central city, a peak-only ‘secondary spine’ (Modelling Report, p 42). GW had finally recognised that the golden mile route is overloaded. Later, the Spine Study was quietly abandoned, and with it the secondary spine. The golden mile now carries fewer buses than it did this time last year, but new double-deck buses, notorious for stop delays, have frustrated improvements.

    The business case for light rail is this:

    — Inner Wellington has only three north-south streets suitable for public transport: the waterfront (Waterloo, Customhouse and Jervois Quays), Featherston Street and Lambton Quay. That is a total of some ten lanes, plus parking, for all purposes.
    — The golden mile is badly overloaded, making it a major source of bus delays and missed connections.
    — Overloading is so bad that both golden mile and any relief route will be at their free-flowing capacity (around 40–50 bus/hr each way) as soon as the relief route opened.
    — There is no space for a third public transport route in the central area, so the second route must have enough capacity for future public transport ridership growth. That means heavy rail, BRT or light rail.
    — Heavy rail would have to be grade-separated almost throughout: possible but noisy and very costly.
    — BRT as chosen in the Spine Study was useless, because it had no more capacity than a conventional bus lane. Again, it was quietly abandoned after the Spine Study. Any new BRT project would have to meet at least the ITDP ‘Bronze Standard,’ as identified by LGWM’s consultants. That would require four lanes at stops, to avoid minor bus delays affecting other buses, and Wellington has no space for it. Brisbane’s BRT stops are about twice the width of Manners Mall.
    (Rubber Tyred Trams are just another form of BRT).

    That leaves either light rail or the present-day muddle, continuing its slow descent into chaos as the city grows. The rest of the light rail case can easily be sorted out by LGWM and their consultants.

  22. luke, 10. February 2019, 18:02

    people need to accept transfers are an integral part of networks. The transfers need to be fiscally non penalizing and frequent. Even a heavy rail extension or tram trains would incur a transfer unless every waikanae/hutt/jvil train was running right thru which would require multiple tracks.

  23. Mike Mellor, 10. February 2019, 18:49

    There’s generally a basic difference between road proposals and rail/public transport proposals in urban areas: the former are intended to bypass key urban areas, the latter to link them.

    Being a road project (the NZTA public transport proposals were clearly an afterthought), the Mt Vic tunnel-Ruahine St route bypasses Hataitai and Newtown. That is good for traffic flow, but the opposite of what is required for public transport patronage: buses and light rail need to serve major centres directly, not avoid them. (There are exceptions, but they are generally for longer flows such as the Northern Busway in Auckland following SH1.)

    That’s the reason the main eastern suburbs bus routes do not go along Ruahine St, with its minimal transit potential, but instead go through the centres of Hataitai or Newtown. (Express routes do bypass the intermediate centres or run non-stop through them, but no-one is going to build a new public transport network for peak-hour traffic only.)

    So what is needed a proper public transport route that doesn’t skip the places it needs to serve, the ideal being a high-quality “string of pearls” route through Newtown and Kilbirnie, with a tunnel under Mt Albert making perfect sense.

    A double-deck tunnel under Mt Vic may be perfectly feasible, but that’s not the issue. Just because it could be built doesn’t mean that it should be built – better route options are available.

  24. Roy Kutel, 10. February 2019, 21:13

    Kerry – are you saying GWRC wasted $1million ?

  25. Glen Smith, 10. February 2019, 22:21

    Keith. I agree with a lot of your statements and especially about our planners having fixed plans, failing to think laterally and being unable to undertake long term planning. One of the first slides in my Basin presentation just had two lines ‘long term’ and ‘strategically’. Extra tunnel capacity across Mt Victoria ( the topic of this article) is required due to ongoing eastern growth. Adding tunnel capacity is extremely difficult and so ideally we should anticipate future growth. But on what timescale? 30 year? 50 years? The 80 years you suggest? 100-200 years? Longer?? The logical ultimate is to forecast any growth that will EVER occur (we are talking about a fixed geographical area) and add sufficient capacity now so that across Mt Victoria NEVER has to be revisited again.
    Anticipating growth long term is difficult but since the Wellington population is projected to grow rapidly, there is a move towards high and medium density housing closer to the city, there are large areas of flat land with low density housing in the east and this area is within a short distance of the CBD we can anticipate high growth in the east this century and beyond – in my view much higher than the doubling you predict.
    In terms of adding the necessary additional tunnel capacity, we can confidently say that we cannot achieve this by just adding inefficient road capacity – we need the efficiency of a high quality dedicated PT corridor, ideally rail (each line potential capacity of up to 8 lanes of traffic). The Opus TN24 baseline forecasting report looked at projected volumes across screenlines (appendix A table 1 and 2) and showed peak volumes to and from the east rising to over 150 % of current levels by 2041 (before we might anticipate the main growth in the east over this century). The projected LOS (appendix B) improves briefly with the second Mt Victoria tunnel but rapidly falls back to Level D and large swathes of the city dropping to levels E or F.
    If we are adding rail, is a single lane tunnel (as per FIT’s plan through Mt Albert) adequate? The likely time delay if trains arrive concurrently is up to 1-2 mins. This is unlikely to be a problem initially but as growth occurs and more frequent peak trains are required this is likely to become a logistical problem. We should plan long term and, while we are making changes, add a dual rail tunnel.
    A stacked multipurpose Mt Victoria Tunnel adds 2 lanes of rail, a dual cycleway, a pedestrian corridor and two lanes of road – all within one structure. Will this be adequate to supply the necessary long term across Mt Victoria capacity? in my view yes – probably forever. Without the rail, definitely not. Are there other dual rail tunnel options that can supply the same capacity as cheaply and with as little additional destruction? In my view not. If others feel there are other options then I would am keen to see the details.
    Guy – perhaps you could present details of the separate rail and road tunnels, including approaches, that you feel will be so much easier.

  26. Glen Smith, 10. February 2019, 22:24

    Mike Mellor. Which aspects of my arguments against running rail through Newtown do you logically disagree with?

  27. Mike Mellor, 10. February 2019, 23:32

    Glen, most of your arguments are a matter of opinion, and I beg to disagree. Newtown is no more or less “adequately served by buses” than many other parts of Wellington, and it will become less adequately served as demand increases. A route that avoids the cost and complexity of a double-deck tunnel, avoids further encroachment into the Town Belt, and serves one of the busiest Wellington suburbs – including the hospital – rather than open space and some low-density housing has a lot going for it.

    Of course there will be difficulties: a slightly longer route, but much better connections; adding a need for some transfers but eliminating others; and an opportunity to improve the urban realm through inner south Wellington. Yes, reallocation of road space from cars to more efficient users will be required, but that is a nettle that any sustainable city has to grasp.

    It’s often tempting to make things more complex than they need to be because it appears that that will give a better solution. But the key question is, does the extra initial and ongoing cost of that complexity (whether it’s double-deck tunnels, “tracksharing”, “lighter” units, or whatever) generate at least proportionate additional benefit over the relative simplicity and straightforwardness of a standardised, vanilla solution (which also happens to serve additional major traffic objectives)? All power to your elbow for creative thinking, but often KISS (keep it simple) is a good place to start from.

  28. Katy Mansfield, 11. February 2019, 6:33

    If population is forecast to grow rapidly, Wellington will have other problems to face than just public transport. Water, sewage and electricity being three. Why do we want more people living in Wellington anyway? Why not put all the new arrivals somewhere flat like Palmerston North or Invercargill?

  29. TrevorH, 11. February 2019, 7:43

    The track plans are fascinating but can we have some basic data like: how much will it cost to construct, how many people will use it including projections of the number of trips over 3, 5 and maybe 10 years, and what will the annual operating costs be? We need some yardstick here, even if it has to be largely conjectural at this stage, both to test the project’s viability and to use to make comparisons with other modes.

  30. Keith Flinders, 11. February 2019, 10:10

    Glen. Had NZTA been running things in the 1920s, then the Mt. Victoria tunnel would have been designed to be one lane only, with traffic light control each end I expect. Those running and governing the WCC in the early 1900s displayed a vision which lasted well beyond their terms at the helm. One only has to look at the decision to invest in the tram systems, which opened up Wellington to become a viable community generating wealth and social cohesion. Proportionally the borrowing per head of the then population was massive, but spread over 50 years in the case of the trams, and 90 years for the Mt. Victoria tunnel, the value is evident. So it will be for the next major investments in vital infrastructure, but not vanity projects such as the convention centre. Fast forward 100 years and we see few in power with long term vision. Read more.

  31. John Rankin, 11. February 2019, 11:06

    Glen almost put his finger on what FIT considers is the biggest weakness in its proposal: long-term capacity. However, the weak point is not just as he suggests at the Mt Albert tunnel, but where the line crosses busy roads, because unlike Glen’s proposal, FIT is proposing an on-street line.

    One advantage of light rail over buses is that by giving light rail dedicated lanes and priority at intersections, the service is much more predictable. This means we can schedule light rail operations so that one vehicle never has to wait for another to clear the tunnel. A vehicle will take about a minute to pass through a Mt Albert tunnel. With a planned maximum service frequency of 5 minutes, it is relatively straightforward to stagger the arrivals of up and down vehicles at the tunnel. With stops at both ends (Zoo and Kilbirnie), a vehicle can if necessary be held at the stop for a few seconds while one from the other direction clears.

    However, FIT’s route also requires a tunnel under Mt Cook from Taranaki St to Adelaide Rd. It is not practical to make both tunnels single track. FIT chose to make the shorter tunnel double track.

    In the long term, the maximum practical frequency for on-street at-grade light rail is 3 minutes, maybe 2.5 minutes at a stretch. Higher frequencies generally require grade separation at busy intersections to reduce traffic congestion. In FIT’s estimation, a 3 minute service can operate reliably with a single track Mt Albert tunnel and a double track Mt Cook tunnel.

    As Kerry explained to me: a single track tunnel makes the Zoo and Kilbirnie stops passing loops, at least on one side. In principle, one vehicle consistently goes first, say southbound, then ‘crosses’ (both vehicles at the stop) at Kilbirnie. In practice, controllers will keep an eye on timing, and if the southbound vehicle is late, the northbound may go first: the southbound can always catch up at the Airport.

    If the preferred route for light rail includes a Mt Albert tunnel, FIT expects that council officers will carry out detailed modelling before a final decision on a single or double track is made. However, the big question is how much capacity will Wellington need in the long term on this corridor? FIT’s proposal has a practical maximum carrying capacity of about 10,000 passengers per hour (crush loading could exceed this for short periods).

    Is this going to be enough? Glen makes a good argument that an on-street at-grade light rail line may reach its capacity sooner than we think. Unlike those who question the business case for light rail, who predict too few riders (downside risk), I see the biggest risk is that there will be too many (upside risk). One hopes that the LGWM business case will examine this question.

  32. Bernard C, 11. February 2019, 11:27

    Probably none of those overseas tunnels were built near fault lines. Since they already have a tunnel, it makes no sense to tunnel more and build another one next to it. The way we are now we don’t need to fast forward 100 years! Let’s all recognize the unsolved problems we have NOW (ones the WCC ignores) and look at them. Solve real problems as best we can before we start imagining ones (as the WCC is prone to do).

  33. Jonny Utzone, 11. February 2019, 12:47

    Light Rail will be virtually empty for 20 hours a day excepting Gold Card users who don’t pay. An increase in peak fares wouldn’t go amiss as rate-payers will revolt if they get hit with $400+ rate increases a year to fund LRT.

  34. Casey, 11. February 2019, 13:32

    The strange thing, Bernard, is that all Wellington tunnels came through the 2013 and 2016 earthquakes with minimal damage that didn’t require them to be closed to traffic.

  35. John Rankin, 11. February 2019, 16:31

    Like many others, @JonnyUtzone assumes the purpose of light rail is to serve commuters. Such a line would be difficult to justify in any city. Light rail is only value for money when it also serves all-day, every-day demand. It’s easy to build a light rail line that’s busy during peak periods. The test is whether it’s busy off-peak and on weekends.

    This means serving destinations that have visitors all day, including shopping areas, the hospital, zoo, sports centre, and airport. Many of the people working at these places are shift workers; off-peak services are vital for their travel needs. Peak hour travel is less than half the travel demand in Wellington. City streets are more congested on weekends than during the week. A well-designed light rail line will be busy all day and every day.

    The last few times I rode light rail overseas were to go shopping in the early afternoon, to visit friends on a Saturday evening, and on a Sunday morning. In all cases, the vehicles were full; most of the people did not look like seniors. There is no reason to suppose light rail in Wellington will be different.

    As @MikeMellor put it, “no-one is going to build a new public transport network for peak-hour traffic only.”

  36. Mike Mellor, 11. February 2019, 17:24

    Jonny Utzone: the no 2 bus to the eastern suburbs is currently often standing room only off-peak during the week and at weekends, so why you think that the faster, smoother, more reliable service that light rail would offer would be empty at those times is a bit of a mystery.

    If GWRC embraced the principle of open data we could all see how busy these buses currently are, but for some reason that important information is kept under wraps. That’s costing us all: Transport for London (a much bigger operation, naturally) estimates that the open data it releases boosts London’s economy by well over a hundred million pounds a year.

  37. Graham C Atkinson, 11. February 2019, 18:00

    Mike Mellor – One bus with a capacity of between 56 & 70 (depending on the bus type) every 10 minutes doesn’t equate to a good load for light rail unless (a) the units are relatively small or (b) the frequency is considerably less [neither option very good economics and doesn’t do much for the economic case].

  38. Kerry, 11. February 2019, 19:59

    Graham. You have forgotten about connections.
    — Many bus passengers will transfer to light rail, for example at Miramar or Kilbirnie. This will generally give them a faster trip, because light rail has priority and stops less often.
    — If passengers transfer, fewer buses need run into the city, relieving pressure on the golden mile and allowing much-improved bus services.

  39. Glen Smith, 11. February 2019, 21:07

    John. Why would FIT support an ‘on street’ line (I assume you mean units travelling slowly amongst other street users dodging around it) when a dedicated high quality corridor is possible? By this logic you should be advocating to run the Hutt Units down Ferguson Drive and High Street (how many Hutt commuters would use that??) and the Kapiti Line through Tawa and Johnsonville (ditto). That isn’t sensible planning. Rail excels on high quality dedicated corridors.

    Mike. If we adopt the logical network design of uninterrupted ‘lines’ of the same mode running from a peripheral destination, across the CBD and to another peripheral destination, then Newtown will be adequately serviced by bus since it is on the route for multiple southern bus routes including lines 1 (commonly 10 minute service) and 3 (commonly 10 minute service). What is poorly serviced about that? You intend to unnecessarily force southern commuters to endure a bus to rail penalty to get to the city (around 17 minute ‘disincentive’ penalty) to support your Newtown light rail route. I can see people flocking back to their cars.
    The logical direct route to the eastern suburbs is via Hataitai and Kilbirnie (census populations 6279 and 7280 – not quite “open space and some low-density housing” after all) rather than Newtown (population 8418) while also servicing the Basin, the Colleges, Hataitai Park and Kilbirnie Park including the Aquatic Centre. How many eastern commuters will use rail that takes a detour via the zoo? Once again I can see them flocking back to their cars.
    You claim that running rail via the Mt Victoria Tunnel and Ruahine St adds ‘complexity’ while the weird and wonderful distortions that will be required to run rail via Newtown will somehow be simple. I’d love to see the plans. Even getting down Adelaide Road while preserving parking for businesses, car capacity and the required cycle lane will be a challenge.
    You say that rail via a SH1 route will cost more, without evidence and despite Alun Thomas’s evidence that a dual tunnel is in fact cheaper (around 100 million based on his rough guidelines – you can buy quite a few affordable homes with that).

    You say my statements regarding rail via Newtown are a ‘matter of opinion’. I said the route will be “longer, slower, technically difficult, almost certainly more expensive, forcing all commuters from the East to unnecessarily go via congested Newtown, forcing unnecessary transfers on a significant number of bus commuters and displacing car capacity and parking which will attract opposition from shop owners and residents.” I put it to you these are all facts – I invite you to point out the ones that aren’t.

  40. Jonny Utzone, 11. February 2019, 22:04

    Kerry – just give the buses priority and we could save $1.5 billion and 2 years of construction disruption.

    And Mike, I’m not convinced a full #2 bus with 25% gold card users onboard equals an economical LRT off-peak service. Maybe its because the new GWRC bus service is a tad infrequent?

  41. Mike Mellor, 11. February 2019, 22:19

    Graham: you also haven’t taken account of Newtown traffic, nor of other factors like the service being faster, smoother and more reliable – all attributes that help improve patronage. Current crowded conditions hardly encourage public transport use – but they do make route 2 a relative moneyspinner!

    So JU’s “virtually empty for 20 hours a day” is clearly hyperbole.

    Talk of ratepayers funding light rail highlights one of the inequities of transport funding. An excellent principle is that people who benefit from a project should contribute to it, and road users and taxpayers are significant beneficiaries of public transport projects: for example, they provide a better choice of mode; an insurance policy for when other options don’t work; they take vehicle users and hence vehicles off the road; they stimulate economic activity; and are less carbon intensive. There is therefore strong logic that taxpayers should be part of the funding mix for such projects, and NZTA taking responsibility for Auckland light rail seems to be a move in the right direction

  42. Mike Mellor, 12. February 2019, 0:12

    Glen: your proposal serves the edge of Hataitai – open space and low-density housing, with negligible employment. By contrast, the via-Newtown proposal serves higher-density housing, plus a significant shopping centre including a major supermarket, plus a major hospital, plus a university, plus other medical facilities, all generating traffic to/from both the CBD and the east – an ideal candidate for mass transit.

    I suspect that neither of us are qualified to argue the full detail of what you call facts but I call opinions: for instance, why would a conventional segregated tramway with a single-track tunnel be more technically difficult or more expensive than a grade-separated railway with non-standard main-line-compatible trains through a double-deck tunnel shared with road traffic?

    Properly done, Newtown would not be congested for tram passengers; taking extra land off the Town Belt with major earthworks at both ends of the tunnel certainly won’t be easy; and there’s plenty of evidence that the significance of parking and car access are consistently over-rated by retailers (and it’s likely that parking in particular will become a much less significant issue).

    But enough of this: we’re clearly not going to agree. Good luck with all the bells and whistles with their technical complexity: I’ll go for plain vanilla light rail, serving major centres – much more likely to be fundable.

    Graham: light rail would be faster, smoother, quieter, more reliable and more comfortable than the current buses, all factors that are known to increase patronage. Standing on a late-running (and getting later) diesel bus all the way to beyond Kilbirnie is not exactly a desirable experience!

    Jonny: agree that giving buses priority would be a very good start, and the new bus service is clearly not frequent enough. In the off peak, evenings and weekends the number of buses to/from the Miramar peninsula is roughly half what it was before July, and it shows.

  43. Ross Clark, 12. February 2019, 3:03

    — Many bus passengers will transfer to light rail, for example at Miramar or Kilbirnie. This will generally give them a faster trip, because light rail has priority and stops less often.
    — If passengers transfer, fewer buses need run into the city, relieving pressure on the golden mile and allowing much-improved bus services.

    To comment:
    1. We need to be able to give much more priority to buses now.
    2. Passengers do not like transferring in an NZ context – though I grant you that it would work better for LRT than bus-to-bus – because the overall and underlying service frequencies are not nearly strong enough.

  44. greenwelly, 12. February 2019, 9:11

    Also @Kerry re transfers for LRT,
    Inward transfers are the easy part as LRT is a high frequency service….
    ..But what also needs to happen is an upping of the frequencies of the return bus services. The last thing people want to do is disembark off the LRT and have to wait 20 mins for their “feeder” bus because they just missed the last one.

  45. Mike Mellor, 12. February 2019, 12:22

    greenwelly: dead right, which is why reliability is so important (less important if frequency is high, provided headways are maintained). And there are quite a few scheduled connections that are currently longer than 10 minutes, so there’s a long wait anyway.

    Roll on proper bus priority, pronto!

  46. John Rankin, 12. February 2019, 13:45

    Glen, you ask: “Why would FIT support an ‘on street’ line (I assume you mean units travelling slowly amongst other street users dodging around it) when a dedicated high quality corridor is possible?” You go on to refer to “Hutt commuters” in comparison.

    1. I repudiate your emotive description of on-street light rail. FIT estimates a travel time from the railway station to airport of under 20 minutes can be realised with good design. To achieve this, on-street light rail has dedicated and protected lanes, traffic signal priority at intersections, grade-separation at a small number of busy intersections, and avoids “pedestrian first” streets like the Golden Mile (international guidelines limit light rail to 20 km/hr on such streets).

    2. Your repeated use of the word “commuter” says to me that you see light rail as primarily a commuter service. FIT does not; on the contrary, FIT sees urban light rail in Wellington as primarily an all-day, every-day service, with commuters as an add-on. It needs to be attractive enough that people within its catchment choose to use it instead of driving. The route you propose, as I read it, primarily serves commuters. FIT’s proposed route serves everyone on the corridor. As @RossClark notes above, the underlying service frequency is critical.

    3. Light rail in urban areas fosters transit oriented development around the stops. This city-shaping is a major contributor to the benefits of investing in urban fixed rail infrastructure. It lets more people enjoy the benefits of urban living without having to own a car. FIT’s proposal creates TOD opportunities along almost its entire length. By contrast, TOD on suburban commuter rail lines is conspicuous by its absence. You might wish to comment on the TOD opportunities your proposal creates.

    4. Those wishing to catch the light rail who don’t live within the 1 km “walk corridor” need ways to get to their nearest stop. This is not “forcing unnecessary transfers” but simply a geometric reality. Feeder buses are one option; others include using a bike or e-scooter, getting someone to drop you off, and through buses so you have a choice of a faster trip with a transfer or a slower single-seat trip. Glen, I encourage you to spend 3 months in a city with a well designed and operated transit network, where transfers Just Work. Wellington is an outlier and doesn’t have to be that way. Again as @RossClark notes, we need to start by giving more priority to buses now.

    5. The kind of dedicated high quality corridor you propose typically costs about twice as much per km as the at-grade on-street system FIT has proposed. FIT’s proposal is a “Minimum Viable Product” — more expensive, hopefully better solutions are possible, but there probably isn’t a cheaper solution worth building. FIT’s proposal is mature, proven, and widely-installed, with many successful implementations from which Wellington can learn. I hope LGWM evaluates both kinds of solution before making a final decision. If FIT had the budget that your proposal is likely to need, we’d continue the line north and convert the Johnsonville line to light rail, delivering a faster, more frequent rail service to the northern parts of the city.

  47. Glen Smith, 12. February 2019, 16:29

    John. Glad you are promoting “dedicated and protected lanes” and “traffic signal priority at intersections”- exactly what I believe is the ideal. The description “travelling slowly amongst other street users dodging around it” is an accurate description of exactly where, in my view, rail performs little better than buses. You say you are avoiding “pedestrian first” streets but I wonder why Riddiford St doesn’t fit this description
    I use the word ‘commuter’ (including ‘across town commuter’) to describe anyone travelling by public transport at any time of the day – not specifically to the CBD or work. All day service is essential and I think we both agree that the Airport forms a good ‘base’ for the patronage that is required to justify a rail corridor, since the Airport generates high passenger flows and planes arrive regularly from early morning until late evening. Taking the projected 59,000 daily airport trips and assuming we can achieve around a 25% share (common overseas) would give around 400 passengers per hour in each direction. Therefore the logical destination of rail is the east. If we are going to take rail to the east, then in my view we should get rail to cover as much of the eastern suburbs as possible and I would propose 2 other rail lines, Miramar and Lyall Bay (possibly as a loop line) to pair with Johnsonville and Lower Hutt (based on paired peripheral destinations as part of a seamless regional network using ‘tracksharing’. This would give very wide coverage for rail based TODs in the eastern suburbs. You seem to think that TODs can only be based around rail which displays an unjustified discrimination/bias towards rail.

    The other peripheral area of high population that justifies regular PT services is to the south. Getting rail to the southern suburbs would be extremely difficult – the Southern routes will be bus based for the foreseeable future.

    There are two area of higher density population closer to the CBD – Newtown and Hataitai/ Northern Kilbirnie. One of these major ‘southern’ and ‘eastern’ lines should service each of these areas. A quick look at a map shows that Hataitai/ Northern Kilbirnie fits naturally onto the Eastern rail line. Newtown fits naturally onto the Southern bus lines. Your proposal misses Hataitai/ Northern Kilbirnie/ Southern Mt Victoria on either of these major lines and unnecessarily doubles coverage of Newtown. Despite your protestations, this design just isn’t logical. It seems to be based on an underlying assumption that rail is always superior to bus so we have to run rail to Newtown even if this results in illogical network design.

  48. Paul Clutterbuck, 13. February 2019, 10:16

    The (nowadays ironically named) Tramways Union exists to promote the interests of bus drivers. In the union’s view, the more inefficient a bus service, the more jobs it creates for its members. The Tramways Union is (again ironically given its name) opposed to light rail because it will cut jobs in the sector. It’s a major reason (though unstated) why the union opposed last year’s changes to bus services that were intended to reduce the number of buses on the streets.

  49. John Rankin, 13. February 2019, 16:20

    Glen, some minor comments.

    1. “travelling slowly amongst other street users dodging around it” would be a streetcar, not light rail

    2. Riddiford St is a cars-first street; one test is whether (like the Golden Mile, as Keith proposes above) motor vehicle access could reasonably be restricted to PT and delivery vehicles only

    3. my dictionary defines ‘commuter’ as “to travel regularly, esp. between suburban home and town office” (perhaps the word you want is ‘rider’)

    4. The evidence from overseas, particularly North America, is that TOD happens along urban rapid transit lines; it does not happen on bus lines (reasons include capacity, quality of service, and permanence)

    5. Newtown is a busy all-day, every-day destination and a high-density residential area, so a potential candidate for light rail; detailed ridership modelling will show whether or not it is “logical”

    6. I would really like to see a map showing your proposed lines and stops; can we help with that?

  50. Wellington.Scoop, 13. February 2019, 17:49

    Comments here are now closed, as we’ve reached the maximum number that our system can hold.