Protected Bike Lanes Attract Riders Wherever They Appear

18th chicago 570

pfb logo 100x22

Michael Andersen blogs for The Green Lane Project, a PeopleForBikes program that helps U.S. cities build better bike lanes to create low-stress streets. Second in a series.

The data has been trickling in for years in Powerpoint slides and stray tweets: On one street after another, even in the bike-skeptical United States, adding a physical barrier between bikes and cars leads to a spike in bike traffic.

Now, the first multi-city academic study of U.S. protected bike lanes is out, and a series of anecdotes have formed a very clear trend line: When protected bike lanes are added to a street, bike traffic rises — by an average of 75 percent in their first year alone, for the eight projects studied.

The bike spike showed up at every single facility measured, even those that previously had conventional painted bike lanes.

As pointed out yesterday by CityLab’s Eric Jaffe, these rates of growth met or exceeded citywide bike traffic growth in every case. (The one case in which traffic along the lane tracked citywide bike growth, rather than exceeding it, was Milwaukee Avenue in Chicago, a key route into the downtown Loop that was already one of the most-trafficked bike lanes in the country before protection was added.) On average, bike traffic in the protected lanes grew three times faster than bike traffic citywide.

As the new study from Portland State University’s National Institute of Transportation and Communities notes, about three-quarters of the new users of these routes tended to come from other routes (presumably because the physical separation makes these safer and more comfortable than the alternatives) and about one-quarter of the new users tend to be using a bicycle when they wouldn’t have done so before.

How people bicycling on new protected lanes would have done the trip if not for that infrastructure. Source: PSU

Jennifer Dill, a PSU scholar who co-authored the study, said the evidence is clear that people who bike prefer to do so in these lanes. But it’s not clear yet whether these projects will fulfill their true mission: getting a far larger share of Americans to ride bicycles.

“We’re seeing people who already bike shifting the routes that they’re taking; we’re seeing a small amount of new cycling,” Dill said. “One thing that we don’t know from this or any other research out there is how long it takes for people to really start changing their travel modes.”

On that question, it’s likely to take a few years for academic evidence to come together. In the meantime, though, it might help to go back to the anecdotes. Washington DC, one of the first U.S. cities to start installing buffered and protected bike lanes — and also one of the cities that’s had the most success at getting its commuters to switch from cars to bicycles — now has several years of data on two of its routes. They’ve seen ridership grow much faster than the citywide average, year after year after year:

Peak-hour bike count on Pennsylvania between 6th and 7th Streets NW, Washington DC. Source: DDOT
Peak-hour bike count on 15th between T and Swann streets NW, Washington DC.

Will other bikeways around the country see the same ongoing surge in ridership? We’ll find out.

You can follow The Green Lane Project on Twitter or Facebook or sign up for its weekly news digest about protected bike lanes.

132 thoughts on Protected Bike Lanes Attract Riders Wherever They Appear

  1. And even if it wouldn’t, why wouldn’t you want good biking infrastructure for when you get off the viaducts designed for local travel? You’d need that to get most people to make trips by bike anyway.

    I’ve mentioned multiple times that local protected bike lanes are a great complement to bike highways for the proverbial last mile. You really can’t build one without the other.

    As for what streets to build them on, pick any major arterial-Queens Boulevard (which already has the #7 viaduct for about 3 miles), Northern Boulevard (no room at street level for protected bike lanes even if we wanted them), Union Turnpike (same problem as Northern Boulevard), Hillside Avenue, the LIE, GCP, etc. None of these streets are things of beauty now.

    Access ramps are only needed about every mile or so. It’s possible to make entrance/exits that fit within about 60 feet of street, none of which is sidewalk space. Interchanges with other bike els? Not difficult at all-just make one about 8 feet higher, and run a few ramps. I have no idea of cost estimates, but I would say it can be done for a few billion tops. Spread over 10 years, that’s a tenth of a percent added to the NYC budget. You can even have gates and charge an annual usage fee of $50 to $100. That might actually recoup most of the costs over time.

    Or instead of doing all this, NYC can just get rid of 90% of cars. That will let us get rid of traffic signals and other things which slow down cyclists without resorting to building viaducts. So far we don’t seem to have the political will for this.

  2. Last I checked, Eastern Queens is part of NYC. There are 8 story apartments two blocks from me, along with several good shopping areas within a 10 minute walk.

    Why would we have a larger mode share than Manhattan? That’s easy-because local public transit here sucks big time. The buses are primarily geared towards getting rush hour commuters to/from the nearest subway station. The rest of the time buses run at 15-20 minute headways, plus you often need to change buses 2 or 3 times even on a trip of 3 to 5 miles. 3 to 5 miles is generally too far to walk. End result is many people drive to get around locally, even though most trips are only a few miles. Those are exactly they types of trips which could be replaced by bike. However, that won’t happen until cycling is made safer and faster. Also, I forgot to mention-the roads here are in crappy shape. If nothing else, at least the bike viaducts would mean you won’t hit a pothole large enough to hold a small dog.

    There are other issues as well, such as lack of safe bike parking, but I think lack of safe, fast bike routes is the number one thing keeping local bike mode share here from exceeding that in Manhattan.

  3. Well, roundabouts are a great solution for intersections but NYC seems to have an aversion to them. It’s a pity because they solve many intersection issues without resorting to building expensive viaducts.

  4. Why can’t you have good infrastructure for local trips? Most people have short trips that could be made on bike then. Maybe not all trips would be feasible, but local travel infrastructure is very much beneficial on its own. Your bike highways on the other hand are of limited utility for most potential cyclists without it. So why not build the local stuff first?

    Over GCP means going pretty damn high to go over the overpasses. That isn’t cheap. And not possible by the airport. And just messy by the freight line. That’s a lot of hills that route. Further out it’s less messy, but you still have all the problems of highway median transit. It’s an unpleasant place to be.

    Same with the LIE.

    Union, Hillside, and Northern? Too narrow for on street protected lanes means it’s not wide enough for a viaduct to not be a mess.

    QB would be most feasible of the ones you mentioned, but why build an expensive viaduct when you already have a rapid transit line for travelling long distances? If the line is too crowded, you’d serve way more people building another line to relieve crowding, and per passenger would probably be cheaper since the utility of a single viaduct is minimal without a network to support it.

    Even without traffic lights bikes can’t approach intersections at anything approaching 20 mph safely, typical travel speed through an intersection will probably by 6-8 mph during most of the day.

    Magically get rid of 90% of cars? You mean by building a rapid transit network for people to use instead and destroying low density neighborhoods in the suburbs to put in high density to support frequent service? At that point people will have shorter trips to make so they won’t need your bike highways anyway. Or just move out of the suburbs and into the core.

  5. Yeah, that’s what I like about that solution–and it doesn’t always even need to take up that much space. In many cases standard 4-way intersections with bike lanes can be retrofitted to roundabouts and still have enough space for a cycletrack.

    I know a test Dutch-style cycletrack roundabout with bike priority has been demoed in the UK (one important thing about the demo is it’s been open to the public to try). This design is not currently legal under UK codes but a demo is an important step to getting there. I’d love to see something similar done here:

  6. The suburbs don’t start at the municipal boundary.

    But at short 3-5 mile trips 12 mph is fast enough to make them in a reasonable amount of time so why build the bike highways? Repave roads if necessary, make some roads one way, put in cycle tracks, time lights to 12 mph. Would be cheaper and simpler than riding a half mile to bike viaduct entrance, going up that extra hill, getting off a few minutes later to switch to a different bike viaduct, then getting off, and then riding a half mile to your destination.

  7. Why can’t you have good infrastructure for both long and short trips. You’re only looking at number of trips. I’m looking at the number of bike miles. One long trip could be as many bike miles as 20 short ones. When looked at in those terms, you can see the need for both.

    Look, I get it. You’re ideologically opposed to viaducts or any other type of grade separation except underground subways, so you’ll shoot down any suggestion I have. I might even suggest putting my bike highways underground to appease your concerns but that wouldn’t be feasible on cost or practical grounds in most of the city. Anyway, this is starting to remind me of some conversations I’ve had with global warming deniers. Nothing you say will even get them to consider your viewpoint.

    Moving into the core isn’t possible for most people. Heck, if the house we’re in wasn’t bought 36 years ago and long paid for we couldn’t afford to live ten miles from the core, never mind closer. Most people making less than $200K can’t afford to move where you think they should. If you want to fix that, then either build massive amounts of subsidized housing (the costs of which will make any bike network look like chicken feed by comparison), or create a bunch of $200K a year jobs, even for those without much education.

    Even without traffic lights bikes can’t approach intersections at anything approaching 20 mph safely, typical travel speed through an intersection will probably by 6-8 mph during most of the day.

    No traffic lights means minor streets have stop signs while major arterials have the right-of-way. That in turn means bikes on arterials can safely maintain speed through intersections. Where arterials intersect each other you would have traffic circles. Again, little need to slow down most of the time. You’ll just need to yield to traffic when entering arterials with no traffic signals. That’s a small delay at best.

  8. That’d be nice (and maybe possible in some places) especially in concert with cycletracks but in already-developed areas where do you fit in such a relatively dense grid of off-street bikeways? For retrofitting built environments it simply makes sense to rely on a backbone network of good cycletracks retrofitted into existing street rights of way.

    Don’t get me wrong—fast, longer-distance and even intercity routes are important. But a *lot* of potential bike trips even in low-density suburbia are those local trips under 3 miles to go to the grocery store, ATM, pharmacy, etc. For these types of trips average speed over many miles is not as much a consideration for increased modeshare as much as the convenience/accessibility/visibility (plus, of course, physical separation) you get with separated cycletracks on prominent arterials.

    For these types of trips speed in a car is not even as important…after all most people don’t really time how long it takes to go the 1.6 miles to the grocery store. Sometimes it may be as fast as 3 minutes. Others more like 4, 5 or 6 minutes. But even though 6 is twice as long as 3 it’s not that big of a deal. The reason people drive that all those trips by car is the built environment makes it a no-brainer. Cycletracks on arterials go a long way towards making biking a no-brainer for more people for more such trips.

    Btw, there are some areas that do have roughly 1-mile grids of isolated powerline/creek/canal/etc. bikeways, for example this area of Gilbert, AZ (suburban Phoenix) with the attached image:

    Especially by US standards that’s not a bad network at least on paper. But modeshare is very low probably due to the fact that the routes almost completely avoid commercial areas and suffer from low visibility. An additional network of cycletracks on the arterials could really boost biking for errands and such probably with or without this existing “back yard” network. It’s the visibility thing.

  9. As I’ve said several times already, why not build both? You need the local lanes for that last half mile or so, or for relatively short trips where it doesn’t pay to ride all the way to a bike highway. However, let’s not forget a substantial number of outer borough bike trips will be well over 3 to 5 miles. Back when I used to run errands by bike with my brother, a few times we went 20 miles each way. Sure, we did it on local streets, which incidentally had far fewer traffic signals than now, but it would have been easier if we had bike highways. My take on all this is the way things are now, you’ll only get hard core riders attempting 10 mile or longer trips but bike highways would change that. It’s not so much distance as it is time. Most people will bike to any destination they can reach in 30 minutes or less but few will ride further. Right now in NYC, even in the outer boroughs, averaging 12 mph while obeying traffic signals is a pipe dream. More likely you’ll average 6 to 8 mph. That makes your 30 minute travel radius 3-4 miles. If you had bike highways which could increase the average speed to 15 to 20 mph, your 30 minute travel radius is now 7.5 to 10 miles. Mass produce velomobiles which an average person can afford, run these on your bike highways, and now most riders can go 12 to 15 miles in those 30 minutes.

    You can also do cool things with bike viaducts like roof them over, and channel the prevailing winds into a tailwind. Given the average annual wind speeds in NYC, this last thing could increase travel speeds by well over 5 mph (i.e. your average 13 mph cyclist might be able to hold 18+ mph). My long term goal here would be to put human-powered average travel speeds on par with travel speeds by subway. I think it’s feasible, plus it’ll cost far less than building miles of new subways in places like Eastern Queens.

    Repave roads? That’s even less likely than bike highways around here. The LIE service road has needed repaving for the last ten years at least. They looked like they might finally be getting around to it, meaning they cut a section in preparation for repaving (rendering it totally unbikeable). It’ll probably be like that for months before it’s actually repaved, much like parts of Hillside Avenue last year were a rutted mess for months.

  10. I’m not ideologically opposed to viaducts or grade separation. Just boondoggles. And the blight viaducts bring can be minimized, but it can’t be ignored. It might be worth the cost, but for a bike network, that seems incredibly unlikely.

    You have a viaduct like this in mind? Look at the costs for that tiny stretch. Ramps coming down to street level will take a lot of road or sidewalk space. How do you brush off the impact that has in denser neighborhoods where space is at such a premium?

    A few billion dollars you have in mind for a comprehensive network seems like a pretty low estimate. And even this stretch is estimated at what, $15-20,000 per daily rider. That isn’t horrible, but with NYC construction costs and more importantly serving people who live far out and have to travel longer distances that goes up quickly. (by comparison SAS phase 1 is estimated at $22.5k per daily rider). Not to mention you’ve said you want them wider than this so people can comfortably pass slower riders. Charging someone $50-100 a year would never recoup those costs, or even make a dent in it, probably wouldn’t be worth the cost of collection/enforcement. I’d be astonished if anyone could put together a legitimate proposal that showed that cost per daily rider ten years out was under 100k.

    Brooklyn and Queens alone would probably need something like 300 miles of paths if you want everyone to be within a half mile of one. At grade greenfield paths cost at the high end $5 million per million. Cover the whole city and you’d be spending a few billion at grade with greenfield paths.. Viaducts might increase the cost ten fold or more. Add in NYC construction costs and it gets ugly pretty fast. All this when cycling mode share is as low as it is? Infeasible.

    For a fraction of that you could build out a local access network at grade to make cycling an attractive option for trips of up to 5 or 6 miles for a significant fraction of the population, all the while calming streets and reducing the impact of vehicles speeding by. And for longer trips? People living near trains can take those, add in some new (yes incredibly expensive) rail lines and the need for your bike network goes away. And in the winter you’d have something people like a whole lot more.

    Housing costs are too expensive? Rezone some neighborhoods further in, legalize basement apartments/having roommates etc…Yea, people have to live in a smaller place, but that’s what people living further in now have to tolerate already. There’s no way for everyone to have a large home and an easy commute.

  11. Have the Chinese build it for us. Look how fast and inexpensively they’re building thousands of miles of HSR and subways over there. Maybe when we outsource a few big projects to the Chinese construction costs in the US will come back down to reality. They build double-track high-speed rail lines for something like $50 million a mile overseas and you’re telling me a freaking at grade bikeway will cost 1/10th of that? Are we paving it with gold leaf or something? Viaducts shouldn’t even cost $5 million per mile. I would put them at $1 million a mile as a high estimate. If American construction workers can’t build it for that then we find someone else who will. High NYC construction costs exist because of mob kickbacks, political paybacks, etc. That’s a problem best dealt with by enforcing existing laws. It’s certainly not something we should accept.

    Even assuming the costs for a complete network ended up at $20 or $30 billion, so what? That’s what, $3,000 to $4,000 per city resident spread out over at least a decade. NYC spends over $1 billion a year busing students to school. A comprehensive local and express bike network could eliminate most busing. And then you save on the costs of driving as people switch from cars to bikes. The bike network would pay for itself many times over eventually, just as the subway has. Thank goodness those who lived a century ago weren’t as short-sighted as most alive today or the subways would have been dismissed as a boondoggle.

    It doesn’t matter the distance we’re talking either. Time is time. If you save only 10 minutes travel time per day with a good bike network that time is worth a few thousand dollars each year to each potential user. If 100,000 bike riders each save 10 minutes per day, the value of the time saved annually is in the hundreds of millions, even billions. Does it look like such a boondoggle now? And you don’t think a citiwide bike network would see at least 100,000 regular users? I think it would see 5 to 10 times that as a lowball estimate.

  12. If a Chinese company were to build in the US they’d also face much higher costs due to EIRs, much higher labor costs and regulations that are either nonexistent or sidesteppable in China.

    Not defending high construction costs, per se, just pointing out it’s pretty apples to oranges.

  13. Have the Chinese build it? Seriously, that’s the best you got?

    Yes it looks like a boondoggle. 30 billion dollars, probably still a low end estimate, at 100k riders is 300k per rider. Doesn’t matter how long you spread the costs out for, still damn expensive, more so than any transit project. You ready to spend $10,000 a year to use this bike highway? Only way you’d ever pay for it.

    And eventually you run into issues with bike congestion unless you build these wider, upping the cost further.

    ESA is overbudget like hell. Still costs less per daily rider than what you’re proposing.

    I mentioned distance because per rider costs go up if distance traveled per rider increases.

  14. There might be EIRs, but labor costs should be the same. The municipality pays the Chinese construction firm whatever they quote. They in turn pay their workers whatever wage they normally pay them. As foreign nationals they are not covered by US minimum wage law.

    Totally off-topic, but the long term solution to high construction costs in the US is probably going to end up being robots within about a generation. The US needs plenty of infrastructure work done right now, but just can’t afford it at current rates.

  15. Yeah, have the Chinese build is a reasonable answer when faced with ludicrous construction costs. Also, I don’t get this “overbudget” stuff. I’ve been quoted jobs for work around the house. I don’t want to hear it if things end up costing the contractor more. The quote is what I’m paying. If the company loses money on the job because they lowballed the estimate, that’s not my problem. NYC should do likewise. Get a quote, and then force the contractor to honor it. Either they complete the job for what they quoted, or refund the money in full if they don’t.

    By the way, a one mile grid in NYC would be approximately 400 miles of bikeways but you could probably get by with a 2 mile grid. That decreases the total to 200 miles. I’m still not seeing how even 400 miles will cost $30 billion. That’s $75 million per mile. Unlike that stupidly expensive one-off bridge project, these viaducts would be standardized and mass-produced at a factory, then transported on site to be installed. I’d be surprised if it would cost much over $10 million per mile using this approach. Also, every mile of the 200 or 400 miles certainly doesn’t need to be grade-separated. With careful route selection, you might be able to put half or more of the bike highways at grade. You could probably leverage existing viaducts for much of the rest. I stand by my original cost estimate of a few billion tops. I also think such a system might well see well over 100,000 riders per year.

    Going with your $300K per rider, which I feel is ridiculously high, you’re still well under $1,000 per rider per year if we assume a 500 to 1000 year lifespan, as would be likely if we made these out of recycled plastics or other materials not likely to degrade with time. Unlike roads which see constant heavy pounding, well-designed/built bicycle infrastructure should have a lifespan similar to the Roman roads.

  16. Most high speed rail lines in China are a lot more than $50 million/mile. I don’t have a detailed breakdown of the costs, but if you have a free ROW, nothing nearby to make construction complicated, and all you have to do is rip out some trees, fill in a few spots, lay tracks, and put in some cheap overhead wires? That’s a lot simpler than having complicated interchanges and ramps every mile or so, you’re comparing apples and oranges there.

    You want to finance this over hundreds of years? The interest would cost an ungodly sum. If you borrow $1k now at a measly 1% interest rate after 100 years it’s over $2.7k. Another hundred years it’s at $7.3k. A more realistic 3.47% interest rate (current yield on 30 year treasury notes) is $30k after 100 years, $918k after another 100 years. Financing isn’t free. Don’t pass on massive debts to future generations, finance infrastructure over 30 years or less.

    Did you mean for people to be within 0.5 miles of a bikeway or an access point? One NS bikeway and one EW bikeway? Distance to the bikeway itself is meaningless, distance to access points is what would matter.

    I’ll assume within one mile of an access point to each, NS and EW.

    The city is 300+ square miles of land. Ideally, if each access point served a radius of 1 mile of land that’s 100 access points. If you go back to your original half mile that’s 400 access points. Double it for NS and EW travel. That’s 200 or 800 ramps you need to build (I’m not considering interchanges here) Now each of those is similar to half of a bike/ped bridge like the one linked above. So you’re talking about one-four billion dollars just for access to the bike highway citywide, never mind the highway itself.

    Now you want to build this highway out of some futuristic material that is perfectly smooth for high speed travel (but still allows enough traction for necessary maneuverability at high speeds!?), does not degrade in the harsh conditions of the seasons in NYC, has some complicated roof to channel prevailing winds (would this mean two bikeways, one in each direction, and doubling the number of access points, up to 2-8 billion now without the highway they’d connect to?) and at the same time keep all precipitation out…I haven’t a clue what that would cost or what it would be made out of. Any realistic estimates for it, ideally comparing to something that exists outside of some fantasy design? And the complicated interchanges you would need? What sort of footprint out of them, or rough design or cost estimate? Not to mention the viaduct supports, what are they made out of, what’s their footprint on the ground? I’m having a hard time picturing most of this. How wide would the highway have to be? What sort of throughput would be expected (needed?) on any individual bikeway?

    The more you spread them out the more important safe bike transit on the surface is, and the fewer potential riders on the highway,

    And the thing is, we already have a built network for long distance travel. The subway and commuter rail systems (not to mention highways!) What’s the advantage of building a redundant system for bikes? Wouldn’t it be simpler to just expand bikeshare and put in bike parking so people can ride to a station and then have a bike on the other end of the trip? And to make the trip to the station easier, a nice network of surface level bike lanes, and given the average trip length (no most trips aren’t 20 miles each way like what you do with your brother) it’s good enough to make a lot of trips by bike anyway. Add in autonomous taxis (they’ll be less evil without stressed out drivers) and the need for your bikeway really starts to evaporate. And then you’re left with a multi-billion dollar boondoggle.

    And for the places that don’t have rail? Live somewhere else if you want to be able to get around easier. Sell your home and buy a cheap condo. It’s smaller, but that’s the tradeoff everyone else makes.

  17. OK, we already have highways. Why not then repurpose part of these highways for bikes (and in the process make driving a whole lot less attractive for long distance car commuters)? That certainly seems viable and cost effective. It may even in the end reduce traffic levels enough so we can get rid of most of the traffic controls at intersections which create problems on surface bike routes. Remember we can’t have our cake and eat it, too. A big complaint in NYC (and other large cities) about bikes is that they don’t obey traffic laws. That’s no surprise given how often they would need to stop if they did. Faced with this reality, we can do one of two things-reduce motor traffic drastically so surface bike routes can provide average speeds close to any given cyclist’s cruising speed, or build for cyclists exactly what motorists have had for many decades-namely highways. Cars never would have become as popular as they are if highways didn’t exist as even short trips would take a really long time on surface streets. Cycling won’t become really popular, either, until we make it fast and safe.

    And the thing is, we already have a built network for long distance travel. The subway and commuter rail systems (not to mention highways!) What’s the advantage of building a redundant system for bikes?

    Here’s the reason-accessibility to people of all income levels. Philosophically, I’d be totally opposed to tolling my bikeways, although I might accept a $50 to $100 annual fee as a compromise. That’s still affordable even for the poor. Many poor people with low wage jobs will benefit if they can safely, quickly get to work for $50 to $100 per year instead of paying well over $1000 for mass transit. That’s a huge hit when you’re only making minimum wage (or less) to start with. The entry price of using the highways-namely buying a car and getting a driver’s license, makes them even less useful to the city’s poor than the subways. In the end we need to serve people of all income levels. As much as I’d like to see free mass transit (i.e. paid for entirely via taxes on high earners) that will never happen in the US. However, for a reasonable amount of money we can provide nearly free, fast bike rapid transit to the poor.

    I don’t think a few billion, or even $30+ billion if your cost estimates are right) spread out over a decade plus, is a lot in the NYC budget. It’s a one shot deal which could perhaps be financed by taxes on stock trades and the like. Remember the annual NYC budget is about $70 billion. We’re talking here about adding at most a few percent to that for a decade or so for a major infrastructure project which could radically make things better. Just allowing the poor low cost access to jobs could get many off welfare and into productive work. And then the bikeways would benefit others of all income levels. Good transportation means lots of money saved.

    Additionally, you can’t depend upon subways and mass transit, not when there are frequent delays, or workers who can shut down the entire system by going on strike. A bike rapid transit network provides redundancy in case one of the other modes of long distance transit is unavailable. Redundancy is always a good thing.

    We love to look towards the Netherlands for what they do in cities, but I feel what they do regarding their bike highways ( ) is far more relevant to the conditions which exist in the US than what they do in their relatively small cities. Unfortunately, many SB commenters are either unaware of these types of routes, or just dismiss them out of hand. We’re not getting rid of sprawl in the relatively near future. We may never get rid of it. However, with a connected, comprehensive system of both long distance bike highways and local on-street routes I feel we can vastly mitigate most of the evils caused by sprawl, including the spillover of autos from suburban areas into denser ones.

    At some of the prices we’re paying for rail lines, thinking NYC will expand the subway, as much as it’s sorely needed, is a bigger pipe dream than an elevated bike network. At the prices being paid for the SAS, it would cost the city well over $100 billion to build the 50+ miles of new subways needed in the parts of the city underserved by rail service. Before you start that these are “suburbs”, I call bullshit. Some of these areas are as dense or denser than parts of the Bronx or Brooklyn which have had subway service for a close to a century. Indeed, when the subways were first built out to city limits in the Bronx much of it was practically a greenfield. The development followed the subway line, and it would any new lines.

    I still don’t know where your ridiculous cost estimates come from. In China they’re building subways for $110 million per kilometer ( ). The latest HSR line, admittedly in a sparse area, only cost about $22 million per mile ( ). More typically their HSR costs are about 2-3 times that (the Beijing – Shenzhen line costs about $45 million per mile. That obviously includes a lot of track through dense cities and a lot of tunnels. I’d be surprised if China asked more than $10 million per mile, even for high-tech sophisticated bikeways like I’ve described.

    London is proposing something very similar ( ), perhaps on an even grander scale than what I’m imagining. I’ll look with a keen eye to see what becomes of this idea. It could provide a blueprint for NYC. I’m thinking these things could perhaps even do double duty by providing a place to run electrical utilities along (dispensing with the ugly system of outdated overhead wires which exists in much of the outer boroughs), maybe even pneumatic tubes for small package deliveries (drastically reducing truck traffic). Leveraging them for such commercial uses, they may even turn a profit, giving cyclists essentially free infrastructure while making the city a more pleasant place to live.

    Since you always talk about moving closer in and so forth, consider what will eventually happen when that occurs. Even if you get rid of cars, the streets will be so dense with pedestrians, delivery vehicles, etc, that getting around even by walking will be very slow. Remember subways, as great as they are, really aren’t viable for lots of trips. Faced with the scenario I’m describing, we may need to build separate levels for peds, bikes, and motor vehicles just to keep the city from entering a permanent state of gridlock. At some point you can’t create more space on the surface. We realized that about a century ago when we built subways and skyscrapers. As cities densify further, we’ll be faced with this reality for surface streets as well.

    On a final note, I have some work I need to get done so I doubt I’ll be scanning this thread for replies in the near future. It’s been a great conversation here with lots of back and forth which I hope others found interesting. We may need to agree to disagree on some points, but I feel I needed to make others aware of ALL possibilities here. Sure, protected cycle tracks are really great and I feel they need to be a major part of bicycle planning going forward but they’re not the be-all, end all solution here many believe. We need to carefully weigh the pros and cons of all solutions to build to best solution for any given situation. We also need to lose the aversion to suggesting spending large sums of money where it’s justified. I think that’s the major problem with cycle advocates. We just can’t think big like road planners. That often means the war is lost before we set foot on the battlefield. We need to no longer be content with mere scraps from the King’s table.

  18. I don’t think it works that way to just bring in foreign workers to work on US soil yet they get to ignore US labor and wage laws. Otherwise why would any company have bothered to “offshore” its operations if you could just bring in non-US citizens to the US and not have US laws apply to them?

    Anyway, back on topic it’s pretty clear that cycletracks in built-up areas are way more feasible.

  19. The difference here is short term contract work done by foreign companies in the US versus a US company hiring foreign nationals to work for it in the US. US companies must abide by US labor laws, regardless of who they hire. Foreign companies doing contract work in the US don’t. They could even legally bring in slave labor but I’m sure nobody in the US would want to deal with such a company.

  20. I lived in the Netherlands and think the emerging intercity high-speed bike routes there are great. However, it’s pretty telling that they are only adding these now *after* focusing for decades on all the local cycletracks (still pretty fast, by the way!) + other local measures which really boosted modeshare from all-time lows in the 70s.

    Before you say “yeah, but…” I’ll point out that, yes, the Netherlands is a comparatively small country but you might be surprised at the huge distances people regularly travel for commutes and other things. I only realized this living and working there.

    David Hembrow points this out, too:

    And the truth of the matter is that even with all its sprawl, fully 41% of all trips made in the US are *under* a mile. Yet only a small percentage of those are made via walking/biking. That alone is an insanely huge amount of potential there:

    Though speed is always a factor for all of these under-10 mile trips it really is most important to have convenience, visibility, safety and accessibility. Off-street bike highways can be nice but they’re not modeshare boosters like a cycletrack network along arterials can be.

    Besides, with well-designed cycletracks of appropriate width you can absolutely get good speed in, anyway.

  21. Those are interesting facts. I totally agree you need to build the local cycle tracks either first, or concurrently with the long distance ones. Places like NYC already have a decent amount of such cycle local cycle tracks, which is why I think now the focus should turn to the long distance ones. This is doubly true in the outer boroughs where for the most part a lot of side streets are just fine for biking the last half mile as is, with few changes, but there is nothing really suited for biking longer distances in safety and comfort.

    Obviously there’s huge potential to increase bike mode share for short trips in the US but I think longer trips are equally important. A person driving a mile or two on local errands in the suburbs isn’t bringing their car into the city. Transitioning such trips to bikes, while good for local congestion, does nothing for urban congestion. If we can get someone 5 or 10 miles from the city center to bike into the city instead of drive in, that’s one less car congesting the city. 5 to 10 miles is certainly not an unreasonable distance to bike, but it can’t be done in a reasonable time on most surface cycletracks if the route has many intersections. That’s the key here. If you have intersections every 250 feet, as in much of NYC, surface bike routes just can’t give you good speed no matter what you do. 12 to 15 mph light timing often isn’t viable for motor traffic. Given the prevalence of 2-way east-west and north-south arterials, often any kind of light synchronization, even for motor vehicles, just isn’t possible. And if the cyclist is delayed between intersections by double-parked cars, jaywalkers, whatever, all your carefully planned light timing goes right out the window. That’s really the problem we’re facing, plus safety. If a route doesn’t feel safe, it won’t get used no matter how fast it might be (i.e. not too many cyclists would ride the shoulder on expressways even if it were allowed).

    You also have the congestion. Double-parked vehicles, school buses, pedestrians, food carts, dogs being walked, etc. all reduce travel speed. Nothing much can be done about this, either, so you need to route bike routes away from this mess. Remember, part of what gets people cycling is a stress-free route. From what I’ve seen, cycle tracks on surface streets in places like Manhattan are a clusterf*ck of stress. They may have increased mode share, but at what cost? The predictable ignoring of traffic laws by many cyclists in these routes has resulted in a huge backlash which may in the long run mean less willingness to accommodate bikes as a means of transport. Bike highways would have meant a lot more cyclists following traffic laws for the half mile or so they might be on surface cycle tracks as doing so would present far less of a burden.

    I really think the primary place separated surface cycle tracks would work great and increase mode share is in the suburbs. Most suburban arterials are too hostile to bike. You need separation from motor vehicles. They also don’t have frequent intersections. Indeed, we can even put roundabouts at the one or two major intersections per mile, essentially giving cyclists a safe, very fast route for a very modest cost.

  22. Yeah the point about getting 5-10 (or more) miles into the city (or to the nearest job cluster in a more pluricentric area such as greater Los Angeles or the SF Bay Area) is important, especially in conjunction with local cycletracks.

    Though in the meantime some of this is mitigated by bikes + transit. I live in SF and it’s amazing how my relationship to BART (and thus other connections such as Caltrain) changed once the policy was updated to allow bikes at all times.

    In hilly areas like the Bay Area, transit + bike serves an important role for many reasons, not the least of which is the fact that in many cases it’s impossible to go over a mile (or two blocks!) before hitting a huge hill, much less 10. That being said, a handful of intercity bike routes do already exist here but due to geography something like a comprehensive grid of them is probably impractical in many areas.

  23. NYC absolutely does not have bikeable roads for short trips. If it did, why aren’t people taking bikes on short trips now in any meaningful numbers? What’s different that they’ll tolerate the uncomfortable stressful ride of a mile, or a few miles, to your viaducts, and then tolerate the uncomfortable stressful ride on the other end of their time on the viaduct of a mile, or a few miles, but they won’t take the bike when the destination is only a mile or two away?

    If you want to travel long distance without a car right now, bike to transit. Can’t take your bike with you? Get a folder. In Queens how many people are even 3 miles from the nearest LIRR or a subway stop? Few if any. Turn some streets into one ways, put in a protected bike lane, ban some turns/have separate turn phases from bike green lights, and time lights to 12 mph, and the trip is pleasant and under 20 minutes easy for the people furthest from the train, probably less than 10 minutes for most people. Now how many are currently a pleasant walk or bike ride to a subway station? Almost none. So few would want to make this trip, or even see it as an option. Just as few would see your viaducts as an option unless they live right next to an access point, and have one at the other end of their trip. Don’t want a folder? Citibike from low density housing probably won’t work great, because rebalancing issues would be expensive, but put in secure bike parking at stations in Eastern Queens and have citibike in the denser parts of the city to use at the other end of the trip if needed.

  24. I’ve mentioned secure bike parking at subway stations quite a few times. In fact, back when I commuted to work, and we still had two fare zones, I wished we had such a thing as it would have saved me a lot of money and time over the unreliable bus to the subway.

    Most streets in Queens unfortunately aren’t amenable to any kind of light timing. It’s the crazy pattern of 2-way arterials running N-S, E-W, and at angles in between which make timing lights even for car speeds impossible. That’s actually a big reason bikes aren’t terribly useful out here for longish trips now-miserably uncoordinated traffic signals, plus traffic which is too heavy for much of the day to roll through reds. Cars wouldn’t be that useful for medium length local trips here either save for the fact that you’re usually within a mile or so of a highway. Anyway, given the poor light timing, unless you can regularly roll red lights, you can often walk as fast as you can bike, but you can’t roll reds during most of the day due to the traffic. Not a whole lot which can be done to fix it, either, other than viaducts over arterials, or radically reducing motor traffic. If you start turning streets into one ways, you’ll have problems because often there is no nearby suitable streets nearby going the other way. For example, suppose we made Union Turnpike one way. The nearest east-west street to it once you get past about 220th Street is Hillside Avenue, which is at least half a mile away.

    As things stand now, one way cross streets are bad enough at making short journeys round about owing to the fact that any neat grid pattern here is regularly broken up by parks, expressways, railways, or cemeteries. One way arterials would only make things much worse. However, these exact same natural breaks in the grid represent perfect places to put the bike highways. The portion you would need to elevate to provide non-stop journeys would be well under 50%, possibly well under 25%. Indeed, some bike routes are already mapped out there, only they pass major intersections at grade level. It might be possible to have them go under rather than over given the lack of underground utilities out here. A protected lane ducking down ~8 feet only at major intersections is a pretty low cost way to provide a continuous, safe run.

    The only issue with what I just mentioned is that most of the natural places to put bike highways aren’t near major destinations. You still might end up needing to go on major arterials for significant distances. Last I checked, few of the major arterials here have room for protected bike lanes. Most have two narrow lanes each way, both of which are needed given traffic levels, and a parking lane. Sure, the parking lane could in theory be turned into a bike lane but that will go over like a lead balloon with many residents. Besides, a protected lane wouldn’t fix the problems I mentioned in my last paragraph. These aren’t one-way Manhattan avenues where you pretty much have free reign to set the light timing to whatever you want. The lights have never been able to be coordinated for cars, much less bikes.

    I see a lot of recreational bike usage, so it’s not like people out here don’t consider the streets suitable for biking. Rather, it seems bikes are off their radar as transportation for the same reason they’re off mine-lack of secure bike parking, and very poor average travel speeds during most of the day due to heavy motor traffic. I can actually manage decent average speeds after about 10 PM by rolling lots of red lights, but there aren’t a whole lot of destinations open at that time.

  25. Don’t be ridiculous. Even without timed lights bikes are much faster than walking. Managing 8-9 mph in Eastern Queens doesn’t require too much work on the park of a cyclist. So for trips of under 3 miles, which is a substantial fraction of trips, travel times won’t be too bad. But arterials aren’t safe. And that’s where destinations are. So nobody bikes for transport. Parking is of limited use on its own. You need it to be safe to get to the parking.

    Outside of Northeastern Queens, which while part of the city, is developed more like the suburbs, where they try to minimize the number of through streets, making some streets one way to build a bike network away from cars is possible for the most part.

    If you can’t get any room for bikes for local transportation how is anyone going to get to the viaducts? And biking in the rest of the city isn’t great, so that further reduces the number of cyclists you’ll get. And how are you going to get anyone to tolerate losing sidewalk/street space to ramps to get up to the viaducts? At twenty feet up with a hairpin turn on the ramp you’d still want about a hundred feet to keep the grade minimal won’t you?

    How many people in NE Queens will even want to bike all the way to Manhattan on a regular basis when cycling isn’t safe, or at least isn’t comfortable, for the few miles on either end of your viaducts?

    The 2012 hub bound report had 1748 bikes entering Manhattan over the QB bridge, 1740 leaving, on a typical fall business day. You don’t have a huge number of Queens cyclists you’re building this for. How do you justify the cost, both financial, and in street space you need to take for access points, without a proven large number of users?

    You’re going back and forth on a few things here. Adding protected off street paths, maybe with the occasional bike bridge? Sounds great. One project I would have liked when I lived in Eastern Queens, extending the Vanderbilt Motor Parkway, saw something about that floated recently. Redesigning some streets to add in protected bike lanes? Sounds great. 210th street heading south to the motor parkway by the park and the apartment complex, 209 on the other side. 209 is massively wide. Plenty of room for bikes on that hill, and parking and travel lanes in each direction. Parts of Bell Blvd is two lanes and parking in each direction, never sees enough traffic to need the second lane, plenty of room for protected bike lanes, connects to low traffic side streets and brings bikes to that little shopping district and under the motor parkway, could probably make the connection between the two better. Maybe Kingsbury isn’t bad, but Peck/Richland has a lot of cars speeding through trying to avoid the traffic on Union to get to the Clearview. I used to live right there, my feel was making it not a through street for cars wouldn’t be a terribly hard sell for the people on the street, since they don’t like the cars racing by their homes anyway, and would make the bike connection to 209 motor parkway entrance pleasant to get to from Bell, or they could just pave the other one by 215 instead of just those woodchips.

    My point is, these sort of local bike facilities, off street through parks when possible, occasional bike bridge over the Clearview or some other sunken highway, secure bike parking by transit, better on road facilities? Awesome.

    Building a network of complicated viaducts that aren’t even easy to get to now, for long distance trips, most likely oriented towards Manhattan since that’s where people go? No, a waste of time, money, and advocacy efforts. At least for now. Even if it’s safe to get from your home to Manhattan, how many want to bike that far? I’ll tell you, I’ve probably never gone more than 80 miles in a day, so I’m not any kind of pro here, but probably better conditioned than your typical potential cyclist, and commuting I don’t mind stopping every mile or two to take a breather, and not sure I’d want to commit myself to 25 miles of biking a day. <3 miles to a transit station, or to a store to grab something small, or a bite to eat? That sounds good. And you'll get more people willing to try those trips than the long schlep to Manhattan. Most trips are short. NYC average is only something like 20 miles traveled per person per day, spread over 3 or 4 trips. Most of those trips aren't safe and comfortable by bike, so people drive. Viaducts don't help them much, because of the problems off the viaduct. Occasional protected paths off street have the same problem

    I can't believe there are enough people now to justify your viaducts. I don't know that there will ever be enough people for them to be worth putting in. You know who I think will use them if you put them in now? Recreational riders on the weekends. That doesn't get cars off the road. At most it keeps people from going to the gym.

  26. You want to cut down on motor vehicle use? People have to have another way to get around then. Short trips are bikeable or walkable for most people in terms of distance (<5 miles?). Long trips you'll need to add in transit. NE Queens doesn't have subways? Trains are still less than 3 miles away, usually less than 2, when the weather isn't bad bike to train can be feasible, and good bus service makes it convenient when it rains or snows.

    But then streets like Bell, Commonwealth, Douglaston, 188, Springfield, Little Neck need to have bike facilities.

    And if you want bikes on local trips, you need roads like Union, Hillside, Jamaica, Northern, and Francis Lewis to be safe for bikes too. Travel times aren't vital. 3 miles will still be 20 minutes if it's slow. But more important is safety. Counting the time finding a place to park 3 miles in 20 minutes driving isn't all that uncommon anyway.

    Viaducts can't solve this problem. You still need a way to get road space from cars, and use it for bikes instead.

    Union with a bike lane, some planters, a transit lane, and a general travel lane? Not happening anytime soon. But it would have higher capacity than it does now, so it shouldn't be impossible, and would make for a nicer place to live. Isn't that what we all want anyway? Business owners? You can work something out with them. If I drive to pick up a pizza or something, I never find a spot on Union and have to park on a side street anyway. People in Nassau that want the capacity on Union when the highways are a parking lot? Too bad.

    And since we're talking about a world where large expensive disruptive transportation projects are possible, extend the F down Hillside and increase frequency on Port Washington branch, running them to LIC if you have to, and run frequent buses/LRT on roads going NS from the stations and you give people a good option other than driving to get around.

    Worried about the poor? Instead of building bikeways that you expect them to use to save money you're better off giving them some cash, let them figure out the most efficient way to spend it, and give them enough that you can raise transit fares. There was a link earlier to a Bloomberg View article about reforming the PA, setting it up as some private company mostly owned by the government to try to get it running more efficiently. Maybe the MTA should do the same. Maybe a few private investors pushing will help cut back on some of the union work rules that increase costs of operation, to avoid politicians letting the system fall apart because the city is broke.

  27. On the going back and forth, that’s likely how it would work in the real world. If I were in charge of a major project like this, at first I might give a worst case cost estimate based on the assumption that we need 100% grade separation. I would then need to study it block by block, see if better routing could get rid of some or all of the grade separation without compromising speed or safety. I would also need to look at gradients. It is often worthwhile to go a block or two out of your way if it means avoiding substantial gradients. Seeing the layout of Queens and east Brooklyn, I seriously think we could do “bike highways” mostly at grade, with the occasional bike bridge, or by leveraging already existing grade separated infrastructure.

    When I mentioned suitable streets for biking to/from these bike highways, I’m largely thinking of side streets which are fine for biking as is with few/no modifications. The arterials are where the biggest problem lies. Most aren’t naturally suited to non-stop runs without total grade separation (i.e. viaducts or tunnels). However, we can mostly get around that if the trip can be largely done on a combination of bike highways and side streets. What I’m thinking here is you might only need to ride along the arterial for that last block or two. In that case, it’s perfectly feasible to walk your bike or ride slowly on the sidewalk, and nothing need be done on the arterials. In other cases the arterial may represent the only viable major trunk route for bike travel in that area, and yes, here you probably need to grade separate (Queens Boulevard comes to mind here). However, I really think that will be the exception, not the rule, at least in Queens. Many of the pieces for a great system, such the Vanderbilt Motor Parkway you mentioned, already exist.

    It’s in Manhattan and parts of Queens/Brooklyn near Manhattan where I think my viaduct idea might be most useful. For one thing, the ridership already exists. For another, the streets are crowded, slow, and stressful. Even here, I don’t think we need anything on a scale of hundreds of miles. Consider Manhattan. You can improve the existing east and west river greenways, eliminate existing bottlenecks, remove the traffic lights, or at least keep them green except when a vehicle is driving on a dock, etc. That gives you two bike highways separated by ~2 miles. Stick another one, this one probably done mostly with viaducts, more or less in the center of Manhattan. Add a crosstown viaduct about every mile or so. That’s maybe 35 or 40 miles total of viaducts, and it puts everyone in Manhattan within about 1/2 mile of a bike highway. The protected bike lanes on the rest of the Avenues cover all the last mile stuff. The north south viaducts can connect directly to the East River bridges via flyover junctions. Bingo, that takes care Manhattan. On the Queens/Brooklyn side you have a few viaducts and flyover junctions connecting to trunk routes going east. Most of these trunk routes can be mostly at grade once you get past the heavy congestion near the bridges. Moreover, unlike hypothetical viaducts out near city limits, these viaducts would undoubtedly be heavily used as soon as they’re built. Combined with the (mostly) at grade non-stop bike highways running out to city limits or even beyond, it’ll be possible to safely bike from anywhere in the city to anywhere sure. You’re not expressly designing anything with long distance trips in mind, but in the end, like a highway, they end up being useful for those going only a few exits, or those going a long distance.

    Let me also be clear on why I’m so big on totally separated routes for bikes besides speed/safety. NYC streets have heavy traffic which quickly renders the pavement unpleasant or even dangerous to bike on. Indeed, 99% of my mishaps were due to potholes. Moreover, most bike infrastructure tends to be built on the right where the pavement is the worst. I don’t have any confidence NYC will start keeping its surface streets in good repair given that they haven’t done so in the 51 years I’ve been alive. At least with bike highways which see only bike traffic, once built properly, the pavement will likely remain in great condition for decades, even centuries. Of course, the operative word is “built properly”. You would need oversight to make sure this was done.

    On the biking longer distances thing, I think that’s something we would grow into once people started using bikes for short trips then built the confidence to try longer ones. That’s probably how we all started. That first 1 or 2 mile ride seemed really long to me. Now ten miles is a warmup. So yes, in all likelihood traffic levels on outer borough bike highways might be relatively low in the beginning. They may even take a generation to come into their own. I’ve also little doubt recreational riders will dominate them in the beginning but I see nothing wrong with that. We’re giving recreational riders a safe, pleasant place to ride, just as we give people parks for recreation. In fact, none other than Robert Moses built his first few expressways with parallel bike/walking routes with the thoughts of them being horizontal parks. Unlike real parks, these also incidentally end up being useful for other things.

  28. Mostly good points here. I want to elaborate on why I think building bikeways for the poor is a better idea than subsidizing their subway fares. In the end, the costs of transporting a person by subway, such as labor, infrastructure, energy, etc. end up being higher than the same person transporting themselves by bike. It ends up being cheaper to subsidize biking. Let’s say hypothetically the bikeways give 500,000 poor people in the city a viable alternative to public transit. This number is likely realistic give how many of the outer borough poor endure commutes on slow buses which are only a few miles each way but which can take well over an hour. If we pay for the annual transit fare for these 500,000 people, that’s something like $700 million annually forever. Build the bikeway system once, even if it ends up costing over $10 billion, and you’re more or less done. Sure, there’s annual maintenance, but it’ll likely be in the low millions of dollars. At the same time, since people biking will be fitter, you may well save hundreds of millions annually in Medicaid costs. When you look at the big picture, it totally makes sense to spend a lot money to encourage active transportation.

    The bikeways also fit right in with your idea of containing transit costs. These bikeways would represent serious competition to the MTA. The MTA would have to work a lot harder to sell their service to the public. Either fares would have to come down drastically, or the service would have to be made much faster than biking the same distance. Either way, the public would win.

  29. To be fair, I’ve been at bike photo shoots where the non-cyclist photographer said something like “ride the other way so I can get you and the bridge in the background”. Usually I respond with “no way, that’s illegal!” but I can imagine other models not caring quite so much.

Leave a Reply

Your email address will not be published. Required fields are marked *


As Protected Bike Lane Design Evolves, New Lessons Emerge

Michael Andersen blogs for The Green Lane Project, a PeopleForBikes program that helps U.S. cities build better bike lanes to create low-stress streets. Last year offered lots of case studies for those of us working to make the case for protected bike lanes. With the explosion of protected lanes in the United States, we have far more […]

DC Bike Counts Show Continuing Surge in Protected Lane Use

Michael Andersen is a staff writer for the Green Lane Project. This story was crossposted from the PeopleForBikes Green Lane Project. The older DC’s first two protected bike lanes get, the more spectacular their results seem to become. Freshly compiled bike counts from June 2013 show that the number of people biking in the 15th […]