LRT and transport / land use integration

Proposals to run light rail between Auckland CBD and the airport have been in the news with the Labour-led government saying that it is a priority. So too does Auckland Council, but its not clear who is going to put up the funding.

An interesting question is whether LRT to the airport is a  'pet' transport project in search of some land uses to fund it, or some hoped for land uses in search of a transport system to support them. I am never that sure when plans talk about land use / transport integration they mean  land uses should integrate with the transport system; or does the transport system need to integrate with the land use!

The preferred LRT route goes down Dominion Road, then along SH 20 to Onehunga and across Manukau harbour to follow alongside SH 20 and SH 20A to the airport. This is a transport investment that the Auckland Plan does signal as being helpful for land use intensification along the corridor, so it is not just about getting from the airport into town. In fact when you read the reports, the genesis of the project seems to more about relieving pressure on the central area from too many buses trying to get in and out at the same time. High capacity transit on Dominion Road will take out some buses and it is relatively easy to run LRT into the central area along Ian Mckinnon Drive and Queen Street, compared to trying to bring LRT in from Mt Eden Road or Manukau Road via Symonds Street.  Well, at least that is what I took from scanning the Central Area Access Strategy (note 1)

As one report puts it: "For bus operational and cost reasons it was concluded that LRT would first be built on Dominion Road and Sandringham Road, with other corridors not required until an unspecified later date". At some point the airport LRT link seems to have been added from the bottom of Dominion Road, replacing previous proposals for heavy rail to the airport. Now the project is transforming itself into a city shaping project.

The fact that the transport project is morphing into a land use project is not surprising, given unknown funding. Apparently Minister Twyford has spoken about the possibilities for public-private partnerships, value capture uplift in areas benefiting from rezoning and transport infrastructure investment and/or area-based rating schemes to help fund the project. Some sort of "Special Purpose Vehicle" to finance the project is possible.

Down Dominion Road, LRT will  apparently run in the existing road corridor in the two central lanes, leaving the two side lanes for cars, buses, trucks and cyclists. I dont think there will be any kerb side parking.  At stations, there will be build outs of footpaths so there is only two lanes. Other traffic may just have to wait for LRT services to stop and drop off and pick up people (just like the trams of old) before they pass through. 

LRT is not the only option for access to the airport. Bus Rapid Transit (BRT) is also an option. I think the idea of BRT would be that the buses would run along the outside lanes, which would be dedicated bus lanes 24/7. BRT may not be cheaper than LRT. BRT may require a lot more land along corridors to fit in 'in-set' bus stops  designed so that buses that do not want to or need to stop can get pass buses that are stopped, without the through bus moving into the central lanes.

Which is better from a land use perspective - LRT or BRT? The greater Auckland website listed the following differences between LRT and BRT:

LRT is better where:

• Corridors with particularly high bus volumes and projected future demand
• Corridors serving constrained, high-amenity locations where expanding road space for extra capacity, loading or turnaround facilities is particularly difficult
• Corridors with lots of good opportunities for transit oriented developments and intensification.

Bus Rapid Transit is better where:

• Lower demand corridors and those serving less constrained locations
• Corridors where there is huge value in being able to implement an improvement incrementally and leverage off existing infrastructure
• Corridors with less opportunity for transit oriented development and intensification.

The differences in terms of land uses interest me.

In simple terms, you might say that LRT helps to concentrate land uses along the corridors served. In contrast, BRT tends to support more dispersed land use patterns in the wider area served. The way I like to look at it is like this:  LRT runs on a defined corridor that has a start and a finish. It also uses defined stops. If the LRT is fast and serves many employment, educational and recreational activities, then being close to the LRT line (ie about 400m) is better than living further away from the line. What is more, the fixed line and stations provides certainty to land uses that the LRT will be there for the long term. There is even a theory that due to the investment in fixed assets required, no public agency is going to walk away from the LRT service once installed, even if it is not very profitable to run.

In contrast BRT systems tend to work by a number of bus services from a wider area joining the BRT line at various points. The BRT line is not necessarily the start or the finish of the bus trip. There may be transfers from feeder services to line haul type services, or services may just leave the BRT line and get onto local streets. Either way, the transport benefits tend to be more widespread than LRT (ie less concentrated and more spread out over a wider area). Also there is often a question mark over the longevity of a BRT system. Will bus services be taken off or re-routed at some point? Will the green paint get removed?. This may dampen some excitement over land use benefits.

Of course in both cases, to help stimulate and support land use development, both systems need to provide some sort of transport benefit. It will not make much difference to people’s locational choices if a shiny new LRT service takes just as long as the clunky old bus it replaces, in terms of travel times from A to B. It is also necessary to have some latent demand to unlock. LRT is probably not going to turn around a declining area by itself. At least we do have lots of housing demand in the area.

What either transport option does for the environment along a corridor is not very clear.  LRT sounds good, but the carriages may be up to 66m long. Nevertheless, for some reason LRT seems to fit better with shared space type arrangements than buses, perhaps because capacity is greater, so services are not so frequent. Buses will be frequent and occupy the kerb side lane - never great for pedestrians or cyclists.

If one of the benefits of light rail, compared to the alternative of bus rapid transit, is the support that light rail provides to land use development along corridors, then that extra development may help to off-set some the effects from loss of parking and the like for small businesses along the corridor, ie much more local foot traffic. BRT may not be so kind to local businesses, as benefits to land uses are more widely dispersed.

Bring in funding issues and LRT begins to look better than BRT. Due to the geographical concentration of benefits from the LRT line, it is easier to ascribe a connection between the transport investment and uplift in land values, sale prices, rents and turnover. Getting lots of growth to happen close to LRT stations will be very important if Minister Twyford is going to fund the project by targeted rates, value capture and the like through his "SPiV" (Special Purpose Vehicle).

So a lot depends upon the land use pattern that will integrate with the transit system. If LRT is the preferred mode, then there are some pretty important implications to work through.

As the LRT reports on AT website notes, ‘this area is highly dependent upon the Auckland Council Unitary Plan permitting increases in density in the vicinity of LRT (or BRT) stops. It is imperative that the policy framework supports the delivery of these benefits otherwise projected wider economic benefits will not be realised. Most case studies of transit schemes that have achieved economic uplift have included frameworks for increasing land use density near them. Conversely, transit investments that did not include such frameworks have often failed to achieve an economic uplift e.g. Sheffield Supertram”.

You would therefore think having some sort of idea of future growth potential would be very important to the decisions around LRT. Should the LRT route be looked at more from the point of view of the potential for land use intensification,  than what route is the easiest to stick some tracks down?

It is useful to look at possible LRT routes between Onehunga and the Central Area. There is really only one route from Onehunga south to the airport. In terms of the airport to Onehunga leg, a route beside a motorway is never a great starting point from a land use perspective, but I guess that is the only realistic route?

From Onehunga north to the Central Area, the preferred route is a dog leg route  west along SH 20 then north along Dominion Road. But this is not the only choice.  The Central Area Access Strategy also says Manukau Road could have LRT in the future, but getting LRT into the heart of the central area (through Newmarket and Symonds Street) is a bit too complex, so best start with Dominion Road services. But would Manukau Road be a better bet for land use intensification along the corridor and key points like Newmarket, Onehunga and what is beginning to get going around Greenlane (like the Alexandra Park development)?  A Manukau Road route may be shorter and faster (therefore more of a accessibility benefit) and may be able to support more land use intensification.


As an aside, the Airport to CBD study looked at two different BRT routes from Onehunga into the Central Area (Manukau Road and Dominion Road), but only one LRT route (Dominion Road). This was because previous decisions said that Dominion Road was to be first cab off the rank. Does that feel a bit lopsided?



So at the moment, Dominion Road is the preferred route. Sure enough, the Auckland Plan refresh shows a ‘development area’ down Dominion Road. But is looks pretty spindly.

The following is from the Auckland Plan refresh

A light rail service along Dominion Road would act as a catalyst for development around future stations. The area has a number of established centres, including Balmoral and Valley Roads, with large amounts of mixed use along the corridor. There are good bus routes with high levels of established public transport patronage as well as some cycle connections to the city. There is feasible capacity of approximately 1,800 dwellings which could increase following the completion of light rail.

Mmmmm.. 1,800 dwellings doesn’t sound like much. Of course this is not the only area that will be served. There is Onehunga itself and Mt Roskill / Three Kings is also a likely redevelopment area. But then Three Kings Quarry is being developed now.  The Auckland Plan shows no development area around Mt Roskill. Surely there needs to be a big node at the southern end of the Dominion Road route to help anchor this 'pivot point' on the route?

Benefits of LRT to the development of Onehunga, the largest potential 'node' away from the central area, will be very dependent upon travel time benefits, both north and south. If you were a resident of Onehunga would you be very happy with the dog leg route of the LRT into the central area? You may just stick to the normal bus route.

What do the numbers say? Well this is where it is hard to get a handle on possibilities.

The South-Western Multi-modal Airport Rapid Transit Study (draft 2016) (Note 2) has some numbers in it. This report looked at LRT via the Dominion Road route versus BRT on the Manukau Road route. From Onehunga to the airport, the route was basically the same.

Travel time are listed as follows to / from mid town (the new Aotea Station on the Central Rail Link) to the airport:

• LRT (via Dominion Rd): 38-41 minutes assuming 80km per hour on ‘segregated sections’ of the line
• BRT: (via Manukau Rd) 40 minutes.

I dont know why there is a range for LRT and not for BRT.

There is no breakdown of travel times from Onehunga north, nor any comparative stats of travel time for LRT on Manukau Road, or BRT on Dominion Road.

The Manukau Road route is about 2km shorter than the Dominion Road route, by my rough calculation.  So using average travel times, maybe Manukau Road would be quicker for LRT than Dominion Road (but then there is more ‘road running’ on the Manukau Road route. From Onehunga, the Dominion Road LRT route hugs SH 20 for part of the way and can presumably zip along this section).

Currently,  the journey times for the airport Airbus from the airport to the city is reported to be between 48 minutes and 53 minutes, so a 40 minute trip is a healthy saving and should be reflected in increased demand for housing and business space near the final corridor.

But what about land uses? Which route offers the better set of circumstances for intensification?

The 2016 Multi-modal study has some numbers in terms of people and employment in the catchment of the stations, as of 2013, but the numbers look a bit shonky. Plus it is a static picture.

If you look at the Unitary Plan zonings, and compare the two routes, then Manukau Road looks a bit more positive for on-going redevelopment than Dominion Road. Dominion Road looks like it has quite a bit of Single House zoning  until you get to SH 20, while for Manukau Road, the Single House zoning is more to the side of the corridor  The  map below shows Single House zoned sites in yellow and the two corridors.

At the opposite end of the residential density spectrum, the following map shows Terrace Housing and Apartment Building zoning.

Using my trusty (more like rusty) GIS skills, I have compiled the following table of Unitary Plan zonings for land 1km either side of the Dominion Road and Manukau Road routes, from the edge of the Central Area to Onehunga. The figures are hectares.  I did not count all the different zones, just the main ones. I also did not look at issues of accessibility and walkable catchments to specific stations, I just took a simple 1km buffer either side of the route (not the stations). A more detailed look may say that some land on the southern side of SH 20 is not easily accessible to LRT stations on the northern side, for example. Nor have I turned on the 'Overlays' and looked at what constraints exist like heritage listings and volcanic viewshafts.

AUP Zone	Dominion Road	Manukau Road
SHZ	357.7	252.3
MHS	415.6	265.8
MHU	382.7	152.1
THAB	206.8	237.6
Mixed Use	180.4	204.1
Town Centre	38.6	41.8
Metro centre	0	45.2
Industry	69.7	29.1
Open Space	251.8	315.5

Overall,  the Dominion Road route covers more land than the Manukau Road route, being longer.
But Dominion Road has more Single House zone than Manukau Road. Manukau Road has more Terrace Housing and Apartment zoning , Mixed Use and centre-type zoning - all zoning that supports intensive residential development. The Manukau Road corridor also has more open space zone land along it, something that may be attractive to apartment type developments.

Dominion Road does have a bigger pool of mixed housing zoning, especially its southern leg through Mt Roskill / Three Kings.

 It would be good to correlate the two routes with the land values that I have previously developed in my trigger point blogs.

To the left  is a 'heat' map from Auckland Council's GIS showing land values per square metre around the northern part of the Onehunga to city corridor. This data is based on 2014 valuation data so a bit out of date, but the pattern may still be ok. The redder the colour, the higher the land value. Interestingly, the eastern (Manukau Road) side of the isthmus looks like it has higher average land values than the central part.






So there is some interesting choices here in terms of land use redevelopment. The Manukau Road  route looks like it would support more apartment type development, and while having a smaller overall ‘redevelopment’ footprint along the corridor because of the shorter route, there may be more housing units overall. And these units would better suit an urban form of taller buildings close to the corridor.

Dominion Road’s real benefit for land uses is its southern leg and the options for medium density development along the SH 20 corridor, given its northern section is so constrained by heritage areas. But is there much new capacity in this southern section? LRT may well lift property prices along the Dominion Road route, but will it be enough to support the redevelopment of sites near a major motorway corridor on the south side of a ridge? And what of Mt Roskill area,  are there options for significant rezonings in this area?

And what about the lower income households that may get displaced by any resulting gentrification? Are there going to be positive options for them to stay? Would that gentrification choke off further redevelopment? Is there enough open space to support good quality intensification and redevelopment?

I think the principle of LRT on key arterials  in the central Isthmus is great and hopefully in the long run we are likely to get LRT services on both routes - Dominion Road and Manukau Road. Which one to start with? This needs some serious analysis. I would pick the one that has the best chance of succeeding, in the short to medium term, in supporting and enabling good quality redevelopment along the corridor that can help fund the required infrastructure. To do that, there have to be transport benefits. If the funding loop cannot be closed early, then I get the feeling that funding for more LRT lines will soon dry up.


Note 1: https://at.govt.nz/media/1913570/cap-programme-business-case.pdf
Note 2: https://at.govt.nz/media/1927342/draft-smart-indicative-business-case.pdf

Houses, flats and apartments (4) - the evolution of the middle

A further post on the 'middle' and the resurrection of sausage block flats.

The middle -

medium density in the middle ring of suburbs - has got to get going if Auckland is to stop spreading out. So far I have looked at building permit data which suggest a growing  middle market - townhouses, units and terraces around the 100 square metres of floorspace mark on smaller plots of land. These type of units are more affordable than (non shoe box) apartments or stand alone houses.  But they also represent the most expensive category of house on a per square metre of floorspace basis. As someone said (cant remember who) you get half the size of a  stand alone house for two thirds the price. So some competition and more players in the field may help.

I have also looked at other cities like Montreal that emphasize middle level type densities. They are a typology that can provide very liveable communities. The Auckland Plan refresh also suggests that redevelopment on the middle ring of suburbs is likely and needs to be given a push along. The middle ring doesn't have the heritage constraints present in the inner ring, while middle ring suburbs sit between the central employment  hub of the city centre and the peripheral employment areas to the north and south, so provide good accessibility to jobs. There is also a reasonable amount of open space in these areas and often good access to coastal areas so density around amenity is possible (a market winning strategy).

I know the term 'sausage flats' is not very flattering and may be taken to mean that I don't like them. But that is not the case. The issue is more of how to get the best out of them.

Time to look in more detail at the urban design issues present.  Here it is interesting to speculate a bit on the evolution of infill housing in the city.

If we take an imaginary residential street of 800m2 sections (20m wide by 40m deep) and step along that street with different types of infill housing, then we might get the following. (I know, my Sketch Up skills are pretty limited!).

To the far left is the standard single dwelling on the the large section with the garage at the back. A rare sight these days. Next along to the right is a 1960s sausage block - single storey, perhaps 3 or 4 units occupying about 40% of the site. Generally the buildings were in the middle of the site with a driveway down one side. Third along and we have the single level infill of the 70s and 80s.

Here, minimum densities have come into play. The house at the front may have been the original house with the infill unit added at the back.  Next along (fourth from the left) and minimum lot sizes have reduced in area, so two units can be added at the back. In this case the units are two storeys.

From the street, this transition sees a bit of change in character, but all the housing is kind of aligned to the same grid, being perpendicular to the street. This alignment hides the new density at the back of the sites, but starts to generate neighbour to neighbour issues. However given that development is two storeys at most and spaced out a bit on the site, maybe impacts are not too bad. More dwellings have been fitted in with upsetting too much the character of the street (and perhaps the neighbourhood).

Now comes the fun part. Fifth along, with the removal of minimum densities (but with outlook and open space and landscaping requirements), a two storey block of flats is now possible. The same grain of development is maintained with development side on to the street. My Sketch Up model is not 100% accurate but the basic outline is there. The driveway and outlook areas all sit to one side. Not much of a back yard remains.

What is next in this evolution?

Three storeys is the obvious next step. Take the same 'grid' as the previous two storey development and go up one level. Simple

 At this point, the question arises as to whether the grain of development needs to pivot  so that the main bulk of the building is at the front of the site, parallel to the street, not perpendicular to the street.

This is what the mock up of the last three sites tries to explore.

What I am basically saying is, is  it  time  to make a trade off between more bulk at the front of the site but more open space at the back?

If you occupied the single level house between the two units, which typology better provides for your amenity? The side on block to the left which overlooks all of your site, or the 'end on' block to the right which has less of a sense of overlooking?

If you lived in the flats, which offers the better outlook? Even with a 6m outlook space there is potential for a similar set of flats to go on the single house site to the right of the side on units. The flats orientated to the street would have outlook over the street or the back yard, while the side on units end up with a 'backs-to-fronts' muddle.

Of course there are pros and cons. The flats orientated to the street fill most of the front part of the section. There would be more shadowing of the front part of the neighbouring sites. But the back yard is clear of buildings, so perhaps a sunnier back yard than might otherwise be the case.

Where do the garages go in the flats that are orientated to the street - at the back, or the front? If the front, is the ground floor facing the street all garages? If the at the back, how do you get access?

What would the internal living areas be like in the two different models?

What about trees and 'green space'. With the flats side on to the street there is still potential for trees and vegetation to fill  the side boundaries and for there to be views  from the street of the trees between buildings. In the flats that are aligned to the street, the trees and vegetation at the back is more hidden from the street, but may create more amenity for the occupants and neighbouring sites.

In short, infill and redevelopment has been managed to date by trying to fit in more housing without upsetting too many outcomes - neighbourhood character doesn't change that much when viewed from the street, neighbours don't loose too much of their amenity, while the on-site amenity created is constrained, but not too bad.  Everything gets compromised a bit more each step up in the development 'ladder'.

But is it time to rethink this classic strategy of 'compromise all round' given the prospect of a new round of  more intense infill happening?