Accessing The Airport: Essential Rail Links

Transcript

Peter Thornton:Well, good morning— good afternoon, I think it is actually, ladies and gentlemen. I don't know if you read CAPA's little piece in the program, but they've certainly thrown out a challenge. Where are you, Peter? A big challenge you've put out there in those words. But I hope that by the end of my presentation, you may think it's actually not quite as gloomy Not an irretrievable situation to get it right, and in fact, right now there are quite a lot of government, joint government studies, Commonwealth, State, and investment being made, particularly in the road network, and I'm not going to talk about that so much, but I'm going to concentrate on rail. But just as a point, I had a discussion with the Davies Commission in London about new runways for London, and they said to me their land transport was the single most important factor in their decision-making process. So I don't think we should underestimate that. But let's start with a few key questions, and I'm going to try to address those things. And I think we should note that the job of rail is to move very large numbers of people between high points of— points of high concentration of jobs, points of interest, and places of residence. So what type of rail link does Western Sydney need? to be successful as an airport? And I'm focusing on that issue, as an airport. A lot of other things rail can do, but I'm focusing on that question. Is the existing rail link at Sydney Airport a good model? What proportion of passengers can or should be expected to use rail? And to what markets or centres of population, centres of economic activity should the rail link— should the airport rail link? And I guess the question is, what service parameters are needed to create a successful airport rail. So let's just think quickly about Sydney's airports and the rail system. So Metro Sydney has a large and rather diffuse rail system, but it's focused on the Sydney CBD. And Sydney Airport, as you know, is about 9 kilometres south of the CBD, and it's pretty accessible by road, major and minor road links in all directions except the northwest, and that's now being addressed by WestConnex. So it's also accessible by rail services, all rail services, with not more than one interchange from the network, and that interchange is either at Wolli Creek or at Central. But it's not a dedicated airport rail link, and it operates standard double-decker trains. There are 2 stations that serve the airport, domestic and international, but Wolli Creek is also an adjacent interchange station. Now, the new airport, Western Sydney, is 43 kilometres as the crow flies west of Sydney. It's also about 11 kilometres from the railhead at Leppington. And you can see in the graph here that, in fact, the distances are much, much longer to any centres of population and they radiate all around the airport, unlike the connection that exists between Sydney Airport and the CBD. Let's just think quickly about what provision for rail within the Western Sydney Airport plan. Now, the plan exhibited by the Commonwealth, it's both vague and clear about rail at Western Sydney. And it's vague because it doesn't prescribe how the airport should be served by rail and where from. But it's clear in that it makes very specific provision for 2 pairs of tracks, which enables a very substantial rail service to be provided and a possible 2 stations within the site. So there's absolutely no question that this airport can be served by rail. Now, perhaps just step back a bit and look at the existing Sydney rail network. It was built as an integral part of the Sydney rail network, not a standalone dedicated railway. And accordingly, it incidentally serves the airport rather than it being a prime origin or destination. And it was built as a public-private partnership in which the private consortium financed and built 4 stations and they were paid by the government to build all of the other infrastructure. But it was built on time and on budget, but there were some issues. And part of the issue was that they had to shoehorn some airport stations into obviously an existing operating airport, and that led to what I would call— what I call high friction. In other words, it's not that easy to actually use or travel between the platform And the airport concourse. And obviously, the rolling stock's not particularly well suited to travellers with luggage. And now what we're finding, of course, is that trains are often crowded with commuters. In the early days of the Airport Rail Link, patronage was slow to ramp up. Project failed financially. There's a bit of a lack of clarity in terms of a CBD address. But as was pointed out to me the other day, every CBD station can be accessed. So it's quite good in that regard. So it was a very good engineering project, but as a business, wasn't so good. It's now profitable though because the patronage has really started to ramp up and that's because the other modes, particularly the road mode, is getting very congested around Sydney Airport. So I think there are some lessons there in terms of thinking about how we would plan a new airport rail link to Western Sydney. Now we just have a look at what actually happens. This is the Opal card origin destination data for Sydney Airport. Sydney Airport stations. And I think what you can see there is that the top 10 stations, and that generates 52% of the movements to and from the airport, and they're at Central, Circular Quay, Wynyard, Town Hall, Museum, St James, Kings Cross, Parramatta, North Sydney, and Bondi Junction. So very much clustered quite close into the airport itself. And mostly within about 12 kilometres of the airport. So that's a fairly classic kind of shape for the pattern of access to the airport. Now a few years ago, I started trying to understand what makes a good airport rail link and try to answer some of the questions that I posed earlier, and I've undertaken quite a substantial exercise in data collection and analysis, and I used these airports 51 airports around the world. And I collected a lot of data on things like the road distance, the best road time, worst road time, rail time, rail headway, taxi fare, airport parking, rail fare. And all those factors bear in some, to some extent, whether a passenger chooses to use rail or to use some other mode to access Sydney Airport. Thank you. an airport or travel from an airport. And when you crunch the data, what you see is this. And the first thing to say about it is that the ratio of rail-using airport passengers to the total number of passengers transiting an airport varies enormously. There is no overall rule of thumb. The top 24 airports in that chart all exceed a mode split of 20% and they average about 30%. So, Peter, you've got a question for us. European airports attract links— European airport rail links attract both the highest and highest average mode shares, followed by Asia, Africa, but I only had one site in Africa, and Australia. But North America, and especially the US airports— I only had one Canadian airport and that's Vancouver, which actually does quite well— attract the lowest average and virtually the lowest mode shares. And as you can see, even within any continent, there's enormous variability, and I think that reflects the fact that none of the services are actually terribly identical. They're all a bit different. Sydney's not as good as many European airports, but just about better than every North American airport. And I think the rule of thumb then, or if there is a rule of thumb, that it's that a well-connected airport rail link in a country where people are already used to using public transport ought to be able to achieve about 20% and possibly up to 30% of the market share of the airport's total passengers. And my notes say I'm going to make some observations on key success factors, and rather than go on in great detail about those, I just want to say one thing. It's like a lot of things that we've already heard this morning. It's about creating a service that people want to use, not a service that people have to use. And so it's really important that their experience in terms of the airport itself, the travel on the train, and the destination— and in reverse, of course— is those things really work for people and that they, they want to climb on board and use a train rather than some other mode. So I set out to try to predict the usage of rail to access and egress at an airport and this is kind of the generalised model but as they say forecasting is a black art at the very best of times. So the first problem is to try to predict the airport patronage and where it's likely to come from or want to go to within the city. And the second problem is there's no historical data about how far passengers are prepared to travel to an airport to a new airport where there is choice. There's very little studies that have been done in multi-airport cities. I did get one study from Germany that provided some insights, and I used that. But the idea of this model is that it's the product of the total airport patronage and then the propensity to use rail, and that gives the red curve, if you like, at the bottom, which is the the distribution of passengers and how far they're prepared to go, travel to or travel from to access the airport. So that's the model that I've developed. And I developed that for Western Sydney. And I ran it for Western Sydney. And the figure that I came out with after crunching a huge amount of data, you sort of come down to one number. And that number was about 26%. 26%. Of airport users would use rail to access Western Sydney. I think the reason for that is its location. It's very remote and the other modes, particularly the taxi or Uber modes, are going to be very expensive. But it's a terribly important question in terms of what we do about a rail link to Western Sydney Airport that we decide which way it needs to point. And my analysis suggests that only about 19% of all the rail passengers will be sourced within about a 20-kilometre radius. Now, that's because within that 20 kilometres, I suspect people are going to use cars. That's what my view is. Now, rail has to operate in a very well-defined corridor because it needs to move lots of people between defined points. And it's a massive investment. And so we really need to get it right. It needs to point at and connect to the biggest sources of patronage. Now, in the model, I did do some sensitivity testing, and so these graphs can show you a little bit about how patronage on rail would vary if you change things like the travel time, the fares, and some of the competition fares such as a cab or an Uber. And it's, as you can see, it can vary quite significantly. Yeah. that if rail fares were set at the Opal card rates, then for Sydney CBD and Parramatta, rail passengers might be about 24% to 22% of all passengers. If you drop to 16% like Sydney Airport, then it's about 18% to 20%. If you go to around $30, then the percentage using rail drops to about 15%, 16%. So it's quite sensitive. And similarly, if travel times are based on the existing network, The percentage using rail is in the low 20s, but if you can cut travel times to say 30 minutes from the CBD and 15 minutes from Parramatta, this then— the patronage then is going to leap to somewhere above 30%. And also you can see obviously if the competitor modes are— you have Uber-level fares, Then you're going to really reduce the number of rail passengers. So all those things in combination have an effect. So let's just look at one of the studies that is taking place at the moment. The Commonwealth and the State have a joint study on Western Sydney rail needs, a scoping study. The focus to date, though, seems to me to have been more about rail alignments and not about the calibre of service needed to attract passengers, airport passengers to use rail. And there's a number of things that have been looked at, extending the existing system from Leppington, building a new north-south line, Campbelltown to St Mary's. And that begs the question as to whether it's going to be metro-like we're currently building in Sydney or the existing double-decker rail system. And the problem with that is those 2 things are now incompatible. So if you do have a meeting point, then you've got to change change from one system to the other. So there are a lot of questions, if you like, about what a successful rail link to Western Sydney needs to be and where it needs to go. But according to that study, the preferred links are said to be the north-south link between Campbelltown and St Mary's and an east-west link to Parramatta. Now, to my way of thinking, the possible north-south link looks a bit more like a replica of the Sydney Airport rail link. In other words, it's a link which is essentially a suburban railway that incidentally serves the airport. I don't think it's going to deliver competitive performance to attract a large percentage of airport passengers. That's in my opinion. So let's just focus on Sydney Metro West, which is the project which government is now in the process of planning. So this shows what government is currently thinking about. It's a connection using a metro system to connect the CBD to Parramatta. And I think there are some really key questions for that in terms of how it might be considered as a link to Western Sydney Airport. So, I mean, is it really suitable to be extended to serve Western Sydney Airport? What sort of travel time can be achieved at Parramatta? And then how long would it take to get to Western Sydney Airport? What sort of rolling stock's proposed? Is it suitable for airport travellers? Can an effective terminus be created in the CBD? Because both ends are important, not just the Can the fare be structured that's going to be competitive with Uber or a fast bus? Now, my calculations suggest that on the numbers that are given here, the travel time from the CBD to Western Sydney would be somewhere around about 50 minutes. The question is, is that going to be competitive? So now, if you follow the rail link rail in this state, you'd probably be aware of the debate that's gone on for some time about whether or not metro rail can carry more people than double-decker rolling stock as we currently have. Now, I don't really want to get into that debate too much, but actually just to say that on an apples-for-apples basis, they can deliver both about the same number of people. But there are a couple of really big differences, and that is metro as it's currently configured has about 70% standees versus 77% seated on double-deckers. So I wouldn't say standing would be such a good idea for a longish trip into the CBD after you've sat in economy for 14 hours or whatever you might have done. So I do think though some form of single-decker rolling stock is probably better for accommodating passengers with luggage because we've seen some of the problems that exist on the existing rail link. So it's probably better to have a single-decker system, although I do know of one rail link where they do operate double-deck trains. So let's have a look at a few airport rail links that I've ridden just recently. So some of you may have ridden these as well. So they're all in Scandinavia. I had a little trip just recently, managed to travel on the Arlanda Express, which is a purpose-designed for airport travellers. It's a very efficient service. It's very well located and it's got a very good CBD terminus. There's also an alternative slower, cheaper service. Similarly, Oslo Gardermoen Express. It's basically the same as Arlanda. It's very well located, well set up terminus, very good travel time, efficient service. Helsinki is interesting because it's actually more a hybrid and suburban railway on a loop. When I stood on the platform at the airport, I was trying to figure out which way to go until I realised I could go either way, it didn't matter, they both ended up at Helsinki main station. But when I got to the main station, I found that I was turfed out onto a snow-covered platform, which seemed a little odd actually, and I can imagine given the temperature was probably about -5 and the weather gets a lot, lot worse than that, that being tipped out onto that sort of platform is probably not such a good idea. I think we can do better than that. And then Copenhagen is, is very interesting because it has 2 modes of transport. It has mainline services into Copenhagen Central Station, but from that airport you can also hop on the train that travels to Malmö and onto Stockholm out of that airport. And then in addition to that, they have a metro service, but that's only 13 minutes into the CBD, so if you had to stand, it really isn't such an issue. But let's come to what I think is still one of the best airport rail links in the world. It's 20 years old and it was an airport rail link that was designed to be an integral part of the whole airport development project. And why do I think it's one of the best? Well, at that— at the airport, there is not a single step between the door of the aircraft and the door of the train. And that's what I call minimum friction. And that's the sort of thing that I think passengers— that attracts people to use that particular mode. The train is easier to find than any other mode. The travel time's effective, very competitive. So are the fares. Rolling stock's set up for passenger— airport passengers in terms of seating and luggage. There's check-in available at Central and Kowloon stations. It's worked every time For me, it hasn't— they've never lost my bag. And it's purpose-designed and fully integrated at both ends, and it was done by one agency whose task it was to create the airport and its surface access links. So sort of getting to wrap this up, what's needed to make Western Sydney Airport successful? Look, I think these are the key points, and I don't really want to read through them all, but I might say one of them there I think is You know, an airport station where it's easier to find the rail service than it is to find a cab rank, I reckon, would go quite a long way. And what I would suggest is that if we are going to use the Sydney Metro to extend to Western Sydney Airport, then we could look again at the Hong Kong Airport model because what they have there is effectively a single piece of infrastructure on which there are 2 services. Yeah. There is a suburban service and there is an airport service. And if you're travelling on one, you don't notice the other. So they have them overlapped in that kind of way. So I do think that somewhere in the future, we will also need some form of North of the Harbour link because I think that taps into a populace, if you like, who have a high propensity to use to use airports. So if you want to— if Western Sydney Airport wants to attract that market, it's got to find a way to tap into it. So just to wrap up then, I think that the North-South Link is a bit more about urban and industrial development in Western Sydney than it's about making Western Sydney Airport successful per se. And I think if we can afford that rail link, Then let's by all means have it because it will be useful, but I don't think it's the key thing. As I said, on the face of what we know about Sydney Metro at the moment, I don't think it's the right kind of rail link. Travel times are going to be too slow. I think the seating configuration is not ideal. There are some positives that can be built on. And the important thing is that an extension to Western Sydney Airport would then link the 3 Greater Sydney Commission's key centres— Western Sydney, Parramatta, and the Sydney CBD— with a fast piece of rail transport. If you then provide on top of that interchanging to the existing network, then you allow other people to use it. As I've said, I think Hong Kong is a really good model. I think you could deliver a travel time of 33 minutes. I think you could deliver a lot of seats. And, uh, So I think our task really is to keep on encouraging the governments and those responsible to plan and deliver an airport rail link that's focused on supporting the future development of Western Sydney Airport in addition to its other objectives to deliver rail into a rapidly growing area of Sydney. Thank you.

Copyright policy: All transcripts on this site are the copyright of CAPA - Centre for Aviation. Our reproduction policy is as follows: you may quote up to 400 words of any transcript on the condition that you attribute the transcript to CAPA - Centre for Aviation and link to the original video page. All other use is prohibited. While we aim for 100% accuracy in the transcript, there may be some minor transcribing errors.