Honolulu Metro Rail

Rapid Transit Closer to Realization as Honolulu’s Rail Project Breaks Ground

» $5.5 billion, automated rail corridor is expected to attract 100,000 daily riders once it is completed in 2019.

A week after the Federal Transit Administration recommended it for New Starts funding, Honolulu’s rapid transit project took a step forward today with a ceremonial groundbreaking. The massive scheme, which will extend 20 miles from downtown to East Kapolei once construction is finished in 2019, will radically redefine transport on Oahu, offering residents a true alternative to traffic-plagued surface streets and highways.

Honolulu and the surrounding municipalities — incorporated into Honolulu County — are hemmed in by a geography whose natural barriers make the tropical metropolis practically ideal for fixed-guideway transit like the system that is now being designed. With mountains to the north and the Pacific Ocean to the south, there is little room for the city to expand, so the only place it can go is up. The “Manhattanization” of downtown and nearby Waikiki over the past few decades is representative of this trend. And transit is a popular way to get around — The Bus, the local transit agency, carries 236,000 daily riders, and the city has a transit work commute share of more than 10%, which is the highest of any major city without rail in the United States and about the same as the City of Portland.

Honolulu is not enormous: The city (officially, the Census-designated place) has about 375,000 residents while the island as a whole has 900,000. But the deficit of space means there is no room for expanded roads infrastructure, and the lack of adequate public transit infrastructure operating in its own guideway poses a serious threat to the health of the region. Without better transportation, the city will not be able to densify further. Current decentralization trends, pushing habitation into previously untouched parts of the island, will be unstoppable.

Thus the likely commitment of the federal government to the rail project sometime in the next year or so is good news for Honolulu and Hawai’i as a whole, since the city serves as the state’s economic engine. Of $5.5 billion in construction costs to cover the 20 miles and 21 stations, Washington proposes to contribute $1.55 billion ($250 million in Fiscal Year 2012) — as long as the New Starts program continues to be funded. The city, which introduced a 1/2¢ sales tax in 2005, will cover the rest. Real construction activity will not begin for several more months.

The alignment, which roughly parallels the curve of the south Oahu coast, hits most of the major destinations in the metropolitan area, including downtown, the airport, and two institutions of higher learning (including one now being built). Especially when considering already high ridership along similar routes, the 2030 estimates of 116,300 daily riders do not seem impossible. And relatively short extensions west into Kapolei, northeast to the University of Hawaii-Manoa, southeast to Waikiki, and north into the Salt Lake neighborhood would make the line even more desirable if they are ever funded and built.

Despite the clear need for improved transportation systems in Honolulu, however, the project’s gestation has been difficult. Previous rail transit proposals were cancelled in 1981 and 1992 and a planned bus rapid transit line was abandoned in 2004. The arrival of Mufi Hannemann in the mayor’s office in early 2005, though, brought significant political support for a new rail line. The mayor pushed through the transit tax and won a hard-fought election against a rail opponent in 2008, as well as a voter endorsement of the project. A fight with Governor Linda Lingle, who argued that the project was too expensive to justify its costs, ensued.

Yet the recent election of Governor Neil Abercrombie and Mayor Peter Carlisle, both of whom assumed office in 2010, represented a major step forward, as each have been solid defenders of the project. As construction moves forward, the city will benefit from this show of support from the municipal and state governments.

It is true that the project remains under debate on both aesthetic and land use grounds.

The elevated nature of the system has a number of advantages: It will allow trains to run much more quickly between the ends of the island (at almost 30 mph on average) than would be possible with an at-grade light rail corridor running through intersections, and it will offer automated trains, allowing high frequencies even off-peak (6 minute maximum) and lower labor costs because of the lack of train drivers.

Nonetheless, the elevated guideway will not be a particularly beautiful addition to the Hawai’i landscape, and in some places it could represent a barrier between the city and its waterfront. The alignment will require 20 residences and 66 businesses to be bulldozed. It is also expensive: A ground-level light rail line or a busway could probably be built for fewer funds. Yet neither would provide the kind of mobility benefits the automated rail line would.

Moreover, opponents of the project suggest that its appeal — fast transit times from downtown to the far west side of the island — will encourage sprawl in areas around the planned university and in Kapolei. Indeed, there are already proposals on the books for a giant project with thousands of homes that will shift patterns of house-building activity to this area. Is it worth paving over now-agricultural land for the purposes of building park-and-rides with the assumption that in the future these areas will become transit-oriented cities of their own?

But Doug Carlson, writing on his site, poses a different question: Does Honolulu have any choice? Given that the city will continue to increase in population, the number of automobiles running up and down its highways will only ramp-up as well. Assuming that growth is inevitable, the city might have no option but to promote new communities designed for commuting by public transit. In that case, this rail project seems completely justifiable.

56 replies on “Rapid Transit Closer to Realization as Honolulu’s Rail Project Breaks Ground”

Great to see this moving forward.

Here’s a thing for your readers to ponder. Look at that Salt Lake “extension” on the map, and ask yourself, “is it fair to call that an extension?”

The word is technically OK, but there’s a connotation problem. “Extension” sounds like “continuation of the current wonderful thing into new areas.” But branches divide frequency, so a Salt Lake “extension” can’t possibly be that. Neither, for that matter, could the two extensions in the east, unless of course only one of them is built.

“Extension” is a tricky word. There’s stuff hiding in it that can mislead.

But yes, a great project.

I don’t really have a problem with calling the two eastern branches ‘extensions’, but have to agree that the Salt Lake branch is deserving of some qualification. It will certainly be interesting to operate that, even with full automation…

The question in my mind though are (aside from frequency, which will hopefully be addressed by the line’s automation) is what does it do for through passengers, and more importantly, how will this information be conveyed to passengers? Which route should you use, does it matter? If it does, is that always the case? Will any attempt be made to inform people at all?

It’s almost certain that none of this has been addressed yet, but in my opinion it really should be looked at early on, certainly before any commitments are made to building the branch (really it should have been done before approving a plan including it, but what’s done is done).

Otherwise it is great to see a second (full scale) North American example of a fully automated mini metro type system built, they really do seem to have fallen by the wayside with the light rail boom and as Vancouver will attest have some real advantages.

Easy solution:

The station overhead signs will just say “East Kapolei via Airport” or “East Kapolei via Salt Lake”.

The signs on the fronts of the tains can say the same.

Well, The Salt Lake line was one of the two possible alignments between Aiea and Kalihi. For a time it was the chosen path until the City Council decided that the Airport Line would produce more riders, so that is being built first. There are still Salt Lake backers who want a line built through their neighborhood. It is presently served by the 3 (Kaimuki – Salt Lake) bus and the less frequent 32 (Kalihi – Foster Village) bus.

Honolulu may not be enormous but it is supposedly the city with the highest residential density in the country – higher, even, than New York. The line will be a slam-dunk here.

Yeah, Honolulu is very dense – it has the fifth or sixth highest weighted density in the country, if I remember correctly. It also ranks high in transit use by US standards, 8%. But I have no idea where you heard it’s denser than New York.

One link here:

Brookings’ highest density city, Honolulu, is not your image of urban jungle, but rather a tropical paradise jungle! Yet it is 50% denser than number 2, LA. Why is this romantic city the densest? Development and parks have filled the flat areas, which are bordered by steep mountains and the ocean on three sides, and military bases to the west. The nearest substantial undeveloped flat land is a long winding drive or boat ride away. The oversized McMansions which scar much of the countryside elsewhere are not in vogue here; most housing consists of small apartment buildings or single family housing with modest yards. Consequently, people gather in public parks and other public places, generating a sense of community. The density also leads to higher bus use. Honolulu probably averages 5 to 10 households/residential acre, depending on the fraction of land nonresidential and the family size. So it averages above the sprawl range (1 to 5 hh/res ac).

Very excited to see this moving forward!

As far as future branches are concerned, would it make sense to essentially have two “lines” running mostly on the same track? For example, have one line that runs from Kapolei to UH Manoa via Salt Lake. This would essentially act as more of a “residential” line to connect people to work centers such as downtown & Ala Moana Center. The second line could run from Pearlridge Center to Waikiki via the original airport track. This line would act more as a tourist line since it would connect Aloha Stadium, Pearl Harbor, the airport, downtown, and Waikiki. Obviously the main core of Honolulu would be served by both lines. Outside of the core, frequencies could be around 10 minutes, while the main core line (and Pearlridge) would have frequencies of 5 minutes.

It’s about time Hawai’i started building a rail-based mass transit system. The geography of the city is perfectly suited for it. It makes MUCH more sense than that wide “interstate” highway that cuts through the city!

With land at such a premium on the islands, it would seem a true subway might not be a bad idea. An elevated line is not the most esthetically pleasing option, but may be the only viable one. Not an expert, but I don’t think there has ever been any thought to running it underground. The “rock” in Honolulu is nothing more than an extremely porous coral and would below the (salt) water table.

The renders of the new elevated segments look acceptable, but not great:

A good portion of the 1992 Rail Line was supposed to be underground near the State Capitol. However the cost probably sunk the project.

Also worth noting that there is a ton of elevated highway along this same route, so the rail’s not setting any kind of precedent here. There’s a long stretch of double-deck freeway near the airport among other places.

Having just visited Honolulu, I can say that this project is desperately needed. The linear layout of the city along the coast means that a few roads handle all of the traffic, and they get extremely backed up. This means that good transit is a no-brainer. A trip that takes 15 minutes at midnight can easily take 90 minutes during rush hour. It’s pretty crazy. I’ve sat in it, thinking, oh my god, I would have taken transit in a second if it was available. The large number of tourists will quickly and naturally use transit, making this line an incredible success. Very exciting!

I see there are station names for Aloha Stadium, Pearl Harbor and the airport, but how well or directly will the rail line serve them? Are these direct to the stadium, Pearl Harbor memorial, and airport terminal or will they require long walks or bus connections?

The airport station is right there. It’s about as close to the terminal as (for example) the Orange Line station at Midway is in Chicago. The Pearl Harbor station is meant to serve the naval base, not the memorial.

As for the USS Arizona memorial at Pearl Harbor and Aloha Stadium, the two are close: about 0.6 miles apart. If they had wanted to be REALLY tourist friendly they could have put one station at each; instead they split the difference, with the station about 0.2 miles from the stadium and 0.4 from the memorial. The 0.4 mile walk is not what I’d classify as pleasant; it’s HOT in Hawaii, and there’s no shade trees, there’s also not much of a sidewalk and absolutely nothing to hold your interest along Kamehameha Hwy (which can be as many as 9 lanes wide at intersections.) Hopefully they plan something to make the walk more pleasant, like a dedicated pedestrian path closer to the harbor.

Makalapa Park is situated between the station to the northeast and the Memorial to the southwest. Seems like they could do something through that park to create a relatively pleasant walk. The 0.4 mile distance between the transit station and the Memorial should = a 10 minute or less walk.

Following the Vancouver analog, does Oahu have an agricultural land reserve? This seems like the obvious solution to prevent unplanned and unwanted sprawl.

Honolulu seems to have a very unique project on their hands. Their line is basically going to function as “Interurban” rail transit with one major caveat – the planned urban center on the western end of the line will not be built out for decades.

Still, a good point of comparison is Philadelphia’s SEPTA Norristown High Speed Line. It shares characteristics like full grade separation, high frequencies, and third rail power. For transit geeks, the line was the first in the US to use aerodynamic rail cars that had been designed in wind tunnels and optimized for maximizing speed.

Two things that the Philadelphia Interurban line does that Honolulu might want to investigate:

1) It runs a variety of Expresses and Limiteds during rush hour. These trains systematically skip stops during rush hour offering maximal transit times to the end destination. Such an approach may be helpful for further reducing the transit times in the Honolulu rail system.

2) All but a handful of stops on the line are flag stops, where the passengers on the train need to “pull the cord” bus style to signal a desired stop. Passengers on many of the secondary platforms have to push a button on the platform to get a train to stop. The net result of such a system is that no dwell times are incurred when no passengers are exiting and entering a given train.

Neither of the ideas you propose has any place in a rapid transit system, characterized by high-frequency local service and a simple service pattern. Honolulu can expect every train to need to stop at each station; complex station-skipping moves reduce capacity and frequency, and are a feature of very long lines that have passing sidings, and steam-era American commuter lines.

Neither of the ideas you propose has any place in a rapid transit system, characterized by high-frequency local service and a simple service pattern.

A more complex service pattern does not have any place in a rapid transit system characterized by a simple service pattern. That seems to be argument by definition rather than by merit.

Alon, what you are saying is certainly in line with the conventional wisdom, but what I am saying is that there are actual functioning counterexamples.

Clockface schedules may be optimal during offpeak hours, but during rush hours, more exotic scheduling practices can be implemented. This is not theory – it has been implemented for decades on this line. During rush hour headways on this line can be as low as 4 minutes!

And no, all stations do not require a stop unless passengers on a given platform are entering and exiting the system. Aloha Stadium is a necessary station on the Honolulu line, but at any given time during normal commutes it will have very few passengers. Certainly those few passengers should be accommodated, but the next train that passes 5 minutes later should be able to sail through the stop if nobody wants to exit or enter the train.

I know such an approach seems a bit weird, but it has worked on the Norristown High Speed Line for decades. It is a unique transit system, but it does have certain similarities to the Honolulu system that is worthy of study by Honolulu planners.

I wouldn’t really call the Norristown line “functioning.” Wikipedia tells me its daily ridership is 10,000, which is trivial by rapid transit standards. Honolulu is expecting ten times that ridership. The closest thing in North America to what it’s building is Skytrain, where every train stops at every station, ensuring that every single station gets high train frequency all day.

Quick question: What’s the big deal if it stops at Aloha Stadium and nobody gets off? You save about two minutes, tops, and two minutes will not kill a commuter.

There’s also the issue of delays cascading throughout the system. Let’s assume that a train stops at Aloha. The next train has to slow down so that headway is maintained, and the train after that also has to slow down. There is no point in slowing down the entire system when you could just factor the stop in.

Plus, I think you’re overestimating technology a bit here. Automated train technology has not progressed to the point where you can call a flag stop and the computer systems immediately acknowledge that. It might work with a driver, but certainly not with a computer.

I don’t think light rail could carry 100,000 trips per day. This seems like the only feasible option. It’s great that they’ll be able to get high frequency from the automated system. This should significantly shape future development.

@JohnWirtz: “I don’t think light rail could carry 100,000 trips per day”.

It can. The Blue Line in Los Angeles is the same length as Honolulu’s planned rapid transit line, and it carries almost 80,000 people on an average work day, and occasionally carries 100,000 on really busy days. And Boston’s Green Line light rail system carries 250,000 riders per day on 25 miles of track.

However, Honolulu will benefit from faster and more reliable trips by eliminating grade crossings, and especially by automating operations to allow high frequency all day long.

Yes…but. It’s long been noted that the Blue Line has too many grade crossings, which have invited people to do things like park a sofa on the tracks. If the Blue Line were grade-separated from USC to northern Long Beach, then it would allow significant performance improvement. In this case, grade separation could mean leaving the tracks where they are and building over- and underpasses. Not that I’m holding my breath waiting for it to happen.

All that said, it’s great that this is FINALLY moving ahead in Honolulu. Amazing that there are people who see an elevated transit lane as such an eyesore–how about all that pollution from the H1 (and Kamehameha Hwy), not to mention the roar of traffic on an elevated freeway? It’s not visually perfect, but it allows for major mobility improvement without massive increases in paved area or energy consumption.

I know there are significant parts of Oahu already protected as parkland, but as has been mentioned, I would hope there’s some way to preserve farmland. Isn’t there a provision to protect land on Oahu for watershed purposes?

The best thing about an automated system for Honolulu is that increasing capacity will be easy enough – after all, some automated systems have trains every 90 seconds (about 40tph)

Quick question though: Overhead wires or third rail?

The Honolulu Rail project is going to use the Third Rail electric system. They aren’t going to use any overhead wires or anything.

Third rail is IMHO the better choice in this case. It is more robust than catenary, and for the speeds intended, it is suitable. The weaknesses do not really apply in Hawaii (or have you ever heard of the wrong kind of snow?).

Actually, the major fully automatic systems I am aware of are all third rail.

The Northeast Line in Singapore uses catenary. Catenary is banned on lines that go above ground, but the Northeast Line is fully underground. Obviously, this is not the case in Hawaii, where third rail is the logical choice.

Actually I only know why New York City doesn’t have overhead catenary. The reason is that catenary can be affected by large storms, and since large storms do happen they could destroy the catenary, and it also gives people on the ground a risk of an electrocution.

There’s catenary draped all over New York. The same storms that can bring down catenary tend to have lots of rain, Which then pools around the third rail and shorts it out.

I think he’s referring to catenary for streetcars, which was banned in Manhattan (only I think) as well as a few other cities. I love all the old pictures of CBD’s before phone, telegraph and electric wires were put underground – they really were messy with dozens of poles and hundreds of wires in some blocks.

How fast were the streetcars going that they’d need catenary instead of trollywire electrification?

One serious source of clutter can be line intersections ~ this actually gets worse for trolley buses than for streetcars, since they need both a power line and a return line. That is one of the benefits of the battery / overhead electric hybrid systems being developed. You just don’t put the wires through the intersections.

Most likely, there was never any catenary – it was trolleywire. We’re punting on the distinction because we’re contrasting both with third rail, in the context of rapid transit.

The main reasons why subways very often use third rail is smaller tunnel diameters (which means less expensive). The best example for small diameters are the London tube lines.

From Wikipedia: At about 0830 UTC on December 27, Vamei made landfall approximately 60 km (35 mi) northeast of Singapore, in the southeastern portion of the Malaysian state of Johor. Initially, the Malaysian Meteorological Department classified the cyclone as a tropical storm, though it was later re-assessed as a typhoon at landfall.

They may not get the brunt of typhoons/cyclones but they do get tornadoes. Google “singapore hurricane” – there seems to be some confusion about what the word hurricane means in Singapore – lovely pictures of waterspouts and tornadoes.

I can only vouch that there were no cyclones between 2001 and 2006. I’d also never heard of a tornado while living there – the stereotype was that it was a US-only thing. (It’s not, but nobody in developed countries cares about Bangladesh.)

Given that Singapore is located near the equator (1 deg 22 min N), tropical cyclones there are rare. (Insufficient Coriolis effect at the equator). But, as pointed out, not unheard of.

Yeah, there are ever so often tornado’s in the UK and Sweden (don’t follow other countries as closely, but I would guess in the plains of Russia there’d be some as well).

What’s Bangladesh? Is that what the Bangles are named after or came from?


Quote: “”

At grade/on-street light rail/streetcar can carry over 20,000 persons per hour per direction.

The German city of Karlsruhe is building a subway under the main street to carry the increased traffic on its tram/LRT system.

Due to the massive success of its tramtrain, the main street or Hauptstrass sees 45 second headways, with coupled tram sets during peak hours.

This works out to 80 trams (capacity 250 persons per tram or 500 persons per tram set) per hour in each direction, this gives a nominal capacity of over 40,000 pphpd on the one tram or LRT line.

This type of volume certainly justifies a subway.

I think that the Honolulu taxpayer will get very tax weary after this first metro line is built and of course, with a proprietary metro system, they will be stuck with one supplier and that could hinder future development.

In Vancouver, it was found that SkyTrain was too expensive for the Canada line metro and a conventional metro was built instead, both the SkyTrain ART and the Canada Line are incompatible in operation.

Except that Honolulu doesn’t have an extensive collection of suburbanites who like to a) complain that their communities aren’t getting their fair share of transportation dollars, and b) couch these regional disputes as specious attacks on ATO technology. Would you even be commenting, Zwei, were the good folks in Honolulu proposing to build a conventional el rather than something that resembles your local pet peeve?

Here in Portland, the Milwaukie MAX project (a conventional, mostly-at-grade LRT) is projected to cost $1.5 billion for over 7 miles–that’s around $200 million a mile; cheaper per mile than the Canada Line. And this project, passing through some of the most expensive real estate in the country, is being done for $275 million a mile?

Where do I sign up?

Dude, in the rest of the developed world, they’d be outright embarrassed if an elevated line cost $275 million per mile. One line in Tokyo is being built at this cost, but a) it’s elevated over an existing el in the most constrained part of the CBD, and b) Japanese construction costs are the highest outside the US and UK.

(Note to Zwei: it’s not the technology, it’s the country. Milwaukie MAX is really expensive by non-US standards, too.)


My observation is to note that an elevated line with ATO has comparable costs to a surface line without. The high cost of rail construction here still pisses me off on a daily basis… but that has nothing to do with ATO or SkyTrain.

As to why MLR is so expensive… that I have no idea. The new Willamette River bridge is a key piece of the price, coming in around $300 million or so IIRC; but the rest of the line is garden-variety two-track surface light rail. There’s no project element (besides the bridge) that really stands out.

Zwei, the last I heard, the reason why the Canada Line was built (and poorly so) as a traditional subway was that the BC government had a rigid insistence on a public-private partnership. The Rotem-based group came up with the, uh, “winning” bid. You conveniently neglect to mention that the Canada Line is stuck with permanently limited capacity, because the stations were built to accommodate 2-car trains. It will never be able to interline with Skytrain, which already runs at higher capacity. Moreover, although Skytrain is a proprietary technology, it has served Vancouver very well. You are more than welcome to point to its ancestor, the much-hated Scarborough Rapid Transit line in Toronto, as a costly, perpetually-malfunctioning, inadequate solution to a transit need, but that’s not true of Skytrain. (And I invite you to remember that it’s Toronto Mayor Rob Ford who has insisted on not replacing the piece-of-crap RT with light rail, which would work very well in this context, but maybe someday with subway.) When it comes to any measure of cost or efficiency in Vancouver or the rest of BC, you don’t want Translink or the provincial government anywhere close. They both have a bad habit of destroying functional things (like transit systems and ecosystems). Skytrain works well, is visually unintrusive, and has proved its worth despite Translink’s demonstrated ability to screw things up. Compare this to the Canada Line, which has grossly limited capacity, and will force the eventual construction of streetcar or light rail in the Arbutus corridor, if for no other reason than Translink’s inability to pay for enough buses and bus drivers to meet corridor demand. Now, I happen to think development of the Arbutus line is a good idea, but there is no excuse for it being considered vitally necessary when there’s a full-fledged subway less than a kilometer east and less than two years old. Vancouver’s not densely built once you get south of Broadway, but its theoretically high-capacity subway is proven to be less productive than its theoretically intermediate-capacity proprietary-technology Skytrain.

Drewski, the Canada line stations are designed to be expanded to accommodate three-car trains. If you run three-car trains every two minutes, the system can carry 15,000 passengers each way per hour.

Leave a Reply