Spoorweg now

The first thing we noticed when we arrived at the train station in Delft was the extensive amount of construction underway. So much of the station area was fenced off and in various states of disrepair that it was really difficult finding our way out of the station. We had to scale an elevated guideway over the tracks to get from one platform to the next with all of our baggage, and take a long detour around the station, through the bike parking lot, and  finally out onto city streets. As we learned later that week, the station is undergoing a major expansion called the Spoorzone Delft project. The scope of the project is astounding, especially given the recession and sluggish recovery. Total project costs are estimated at 650M Euro and the work is scheduled to take 10 years to complete. 330 Million Euro has been authorized from the National Ministry of Transport, and additional funding has been secured from the Ministry of Housing, Haaglanden regional government, Rotterdam Metropolitan Area, the City of Delft, and a host of private investors. The project includes the construction of a new 1.5 mile four-track underground tunnel with a new main station and municipal office building above,1,200 new residential units, 400-450 capacity underground parking structure, 9,000 capacity bike parking structure, new access roads and tram platforms, a new park, public squares,and the historic preservation of the 400 year old Windmill De Roos and 500 year old Bagijn Tower. It will occupy nearly 100 acres southwest of the old city center. The concept for the interior has raised more than a few eyebrows.


That aside, the Spoorzone Delft project is an excellent example of transit-oriented development and integrated transportation and land use planning. This massive redevelopment project has so many components and moving parts it’s on another scale altogether. The Dutch rarely construct mixed-use projects within individual buildings on single lots. Vertically mixed ground floor retail with residential/office space above is not very common. Because they plan at the block or multi-block scale, they are – to varying degrees of success – able to agglomerate individual single-use buildings across areas slightly larger to create integrated or mixed area developments or zones. This is similar to smart growth movements in the U.S. to plan at the neighborhood or district level. Affordable housing developments are constructed in a similar fashion. That is, there are no true mixed-income housing projects constructed. So as not to concentrate poverty (of which there is very little to begin with), individual affordable housing developments are mixed with other market rate developments at the neighborhood/district scale. Similar to Portland, the national government maintains a 30% set aside for affordable housing. Revenue is generated through land sales as opposed to Tax Increment Financed revenue.


Regional land use regulations reinforce mixed-use zones by confining all new office space to transit stations/hubs. This is really remarkable when you consider that in the U.S., 60% of all transit trips are work commute trips. By concentrating both residential units and employment at transit stations, the short and long-term costs associated with trips, VMT, congestion, energy, and pollution are all reduced. The local municipalities in the Netherlands, are further able to exert even more stringent controls. Transportation, land use, and environmental regulations dictate what happens where. Financial incentives for transit-oriented development are seldom provided for private developers. An extensive outreach effort and public process is required by law. Residents have multiple opportunities to submit comment on projects, and that feedback must be incorporated in future plans.

The political and economic realities here are very different, but examples from Delft offer a lot of guidance about how to be more strategic with our policies, planning, and investments. High density housing and employment located at station areas gives residents quick and convenient access to very high quality train, tram and bus service.Transit service planning at the Delft Station (or any other station in the regional network) doesn’t focus exclusively on peak commute trips to the Hague nor Rotterdam, but is designed to accommodate all travel needs, all day, everyday (10 departures per hour). The Historic central city of Delft is (nearly) car free, and there are plans to extend that to the new adjacent residential development as well.  It’s already a nightmare driving in the area given its dense urban form, limited roadway capacity, and prioritization of transit, bicycle, and pedestrian travel. On-street parking around the central city costs 2.50 Euro/hour all hours of the day (compared to $1.60/hour in Portland from 8-7). Parking garages managed by the City of Delft cost 2.20 Euro/hour all day long. These are the kinds of goals we need to set and pursue.

It’ll be really neat to return in another ten years to see how things have changed when the project’s complete. I only hope, by that time, we’ll be able to say we’ve accomplished at least of few of these things as well. For more information, visit spoorzonedelft.nl.


This is an experimental traffic calming measure.  It doesn’t have a name yet, so I’m calling it the “Wiebelenstraat” or wobbly/wavy street. There are currently  two of them in the Netherlands. The other is for bikes in Rotterdam. You barely notice anything if you travel at or below the design speed. As you exceed the design speed it becomes increasingly unpleasant (click on photo twice for closer look).

I’m a big fan of brick and cobblestone roads and walkways. I find the sound of people walking along roads and sidewalks paved with brick or cobblestone oddly familiar, maybe even a little comforting, and wouldn’t be surprised if it had anything to do with my fascination with cities, and transportation, and old films that feature the two. The many different types of brick they use in the Netherlands make a variety of pleasant sounds. Roads paved of brick or stone can be arranged in an assortment of patterns and serve as tactile and visual traffic calming measures. Brick is permeable. It has higher reflectivity and lower thermal mass than asphalt.  It only looks better with age. I often wish every city was paved with brick or cobblestone – the pedestrian districts within central cities at the very least – but i’ve always imagined it to be so resource and labor intensive, and couldn’t help but wonder how cities could afford to continue to use it on large scales today.

I recently learned that brick roads in the Netherlands are actually not that much more expensive than asphalt ones. Paul Wiggenraad at TU Delft estimates it costs only about 20-30% more than standard asphalt pavement. This additional expense includes materials and labor, but like so many other things transportation-related, does not account for maintenance. As it turns out, brick is relatively low maintenance, easily replaced, ideal in locations where utilities are buried, and when you consider the cost of maintaining asphalt, begins to seem a lot more cost effective in the long term. With heavy rain, sinkholes may become an issue, but that has more to do with the substrate than the surface material, and happens just as often with asphalt.

It seemed everywhere I went, municipal crews were placing bricks by hand, fitting them one by one.




Here’s another way it’s done. It would’ve been really neat to have seen this in person.


One of the first things we noticed upon arriving in Amsterdam – and eventually, everywhere else we travelled in Holland – was the impressive speed and frequencies of the surface trams. It was amazing to see these trams zip by at 30km/h or more, at times, through some very heavy pedestrian traffic. Every 8-10 minutes, another tram would arrive and depart on schedule, stopping for less than a minute before gliding off to the next stop on the line. Because tram frequencies are so high, you don’t have to bother looking up arrival times. Travel time reliability is exceptional and has a profound effect on mode choice and perceptions of the costs associated with travel. Time tables, transit trackers, and apps are unnecessary even as a visitor! Distances between stops are a bit greater, but coverage is so extensive, and stops and routes are so well planned that transfers seem effortless.

Look at all of these happy transit riders!


In Amsterdam, the trams operated by GVB, are a bit newer and fancier, and it’s difficult distinguishing between  trams and buses from the inside. In Rotterdam, RET trams and buses are in a league of their own. On the inside of every tram and bus is an onboard gps trip tracker. These displays conveniently tell you where you are, where you’re heading, and when you’re expected to arrive there. In Utrecht, there are real-time trip tracker displays at every single stop!



HTM, GVB, RET and other operators all use a similar electronic fare instrument called the OV-chip card. The OV-chip card is a convenient, prepaid, reloadable card that you scan at magnetic card readers upon entering and exiting the trams and buses. You simply tap on and tap off. Fare schedules for each operator vary slightly, but to give you an idea, HTM charges a fixed base fee for each trip, 0.83 Euro, and a fixed price per kilometer, 0.139 Euro/km. Also available are floppy passes for 7.50 euro which are valid for 24 hours after the first scan, and several other ticket options. GVB charges a fixed fee of 0.83 Euro and 0.143 Euro/km. They offer disposable one hour passes for 2.70 Euro and 24-hour passes for 7.50 Euro. Yes, this is a fair bit more than what we pay in the States, but hold on! When you consider the level of service provided, its actually a screaming deal. Also consider this: HTM farebox recovery rate is an astounding 60% (230 Million Euro), compared to just 20-25% for TriMet. In principle, higher returns (in addition to relatively higher subsidies to begin with), mean even higher investments in service, resulting in even higher quality service, yielding yet higher returns on investment for everyone. It’s magical.

In what other ways might we account for all of these miraculous accomplishments, and how does Portland compare? A few fundamental factors related to operations and organization set the Dutch apart as having a truly world class transit system.

Signal preemption

In the Netherlands, trams are almost always given exclusive right-of-way and full signal pre-emption. The only exception is when they share the lane with buses. In Portland, the streetcar and MAX light rail are given signal priority. While the MAX has some dedicated right-of-way (along the Yellow Line for example), the streetcar operates in the same space as the rest of traffic, thereby subjecting it to the same delay. What’s the difference between signal priority and pre-emption? With signal priority the signals are merely pre-programmed to extend the green phase when the vehicle is proceeding through the intersection. With signal pre-emption, the vehicle can more or less activate an all red phase for all other modes/directions of traffic at the intersection on approach. In the U.S., full signal pre-emption is typically reserved for emergency response vehicles and the railroads.

President Kennedylann

In the Netherlands, the trams are also almost always center-running with physical separation or given express travel off-street. Buses typically mix with auto traffic and board/deboard curbside, but are occasionally center-running and occupy the same space as the trams. Despite this, I have not seen or heard of corridor specific intercity bus rapid transit in the Netherlands. Instead, bus systems are designed primarily as feeders for the light and heavy rail networks. What’s important is that connections are seamless, and that rider experience is the same, regardless of mode. In theory, center-running exclusive right-of-way means the trams are not subject to the same delay due to stacking or conflicts associated with turning movements of other vehicles. Exclusive center-running right of way and full signal pre-emption – separation in space and time – together, then allow the trams to operate at significantly higher speeds. For the Dutch, these are the keys to making fast, punctual, and reliable service possible. Perhaps the best part about all three of these conditions are that they are all primarily design and technological considerations, which can be adopted in the U.S. We have the technology, and we certainly have the space. What we lack is the coordination, the political will, and the funding (in that order).

We were fortunate enough to meet with two senior planners at HTM headquarters after exploring the tram and bus network around the region. According to Marc Dorst, Beleidsadviseur at HTM, another key difference between transit service in the Netherlands and transit service in America, has to do with the nature and degree of coordination between public and private agencies. The significantly higher level of service that the transit agencies provide can be attributed in large part to this organizational structure. The trams and buses in Haaglanden – the Hague regional area – are operated by HTM, a private firm that is owned by the local municipality, Den Haag (the Hague), and contracts with the regional government, Stadsgewest Haaglanden. By contrast, in Portland, the streetcar is operated by TriMet, our regional transit agency, owned by the City of Portland, and managed by the non-profit, Portland Streetcar, Inc. While both arrangements are a bit complex, the difference is that HTM (and the other transit operators throughout the Netherlands), operate as for-profit private entities whose shares are held completely by the municipalities they serve. According to HTM, this sort of public-private identity advances the basic subsidized provision of service by the regional government, with further incentives to operate and maintain their trams and buses at a level of service that competes with auto trips, attracts high riderships, and in turn enables them to provide even higher quality service. Again, the political and organizational capacity to develop these kinds of incentives, and support continued investment would seem to be our largest and most pressing challenges. A disciplined, coordinated approach toward these kinds of institutional and structural challenges must be reinforced at the same time by the innovative research, design, and policy solutions that are occurring in places like the Netherlands . Only then might webegin to see the light at the end of the tunnel..

So as not to sound entirely hopeless about the situation, check out this clip my friend Ben Weber took the other day of the Portland Streetcar conducting acceleration and braking tests on the eastside. It gets moving pretty quick going backwards on MLK at the end there.


Here is a photo taken from a bike/ped bridge that spans a major arterial with on and off-ramps to the highway beneath it. You can see from this photo a few things that are different about Dutch interchanges. As a motorist exits the main highway on the off ramp (on the left), they are presented with two options: left turn or right turn onto the arterial. They are not permitted to proceed straight through to re-enter the highway (on-ramp on the right), as they might be on other interchange configurations (Note: I wasn’t able to tell if there was a frontage road for the on-ramp, but I’m guessing there was none). This eliminates the need for an additional signal phase,  subsequently improves signal cycle times, and reduces queuing on ramps. It also greatly improves the predictability for other road users including and cyclists and pedestrians. I’m not sure how much this impacts volumes on the arterial, but I imagine drivers soon learn that this is not an option and eventually learn to take the right exit. So simple.

I didn’t get a very good photo of it, but PKoo noticed something unusual about the signal heads for each lane. Both the left lane and center lane become left turn lanes at the next intersection, but the signal heads at this signal seem to indicate the left turn far in advance of the where the actual turning movement is to take place.  It seems a little strange for the signal heads at this intersection to preface the movements occurring at the next intersection, when signage would normally be used for that purpose. Peter, can you speculate/elaborate? The surface treatment (not shown) on each of the lanes that precede this signal also indicate the left turn movement far in advance.

Advanced stop line

Another feature of this design is the super advanced stop line for right turning traffic. The bus has a protected right turn (no right turn on red) with full visibility of bicyclists (and pedestrians) crossing the on-ramp in both directions on the 2-way cycle track which drastically reduces the potential for right turn conflicts.

If there is a SPUI in the Netherlands, I don’t want to know about it!

The last thing I want to talk about on this tour is a remarkable example of traffic calming, affectionately known as “the Bayonet.” It essentially combines horizontal deflection, vertical deflection, and narrowing! If that does not get you excited, I don’t know how you managed to read this far. I took 54 photos of this thing. Behold!

Bayonet (heading north, leaving Delft)

Bayonet (facing north, leaving Delft)

Bayonet (facing south, into Delft)

Bayonet (heading south, into Delft)

The single lanes in each direction on each side of the crossing converge into a single 2-way lane requiring vehicles heading in opposite directions to cross one at a time, yielding to whoever arrived first. The apex of the bridge sits at a height at which oncoming traffic is just barely visible, so motorists must approach very slowly, very cautiously until they can see that it’s clear. Bicyclists, pedestrians, and bus drivers with an elevated line of sight can easily see what’s coming as they approach.

As much as I like this example, there are a few things wrong with this picture. The bike lanes should be cycle tracks given the volumes that this street likely accommodates. Heading north, the pedestrian is asked to walk on the canal side, cross the bike lane, the street, and the bike lane on the opposite side, before proceeding over the parallel walkway. A more thoughtful design would allow the pedestrian to continue on the canal side over the bridge. This could have been accomplished by extending the bridge section of the roadway, and shifting each lane over, thereby making the curve more dramatic, and allowing pedestrians a straight path along the canal. The staging area (box with x) on the north side is awkward, and inconsistent with that of the south side, and if you rolled up on something like this in the U.S. you’d probably need a lot more signage, maybe some flashing yellows, and an audible warning. I wonder if this would ever have a chance of receiving MUTCD approval for experimentation?

There is much more I want to say about all of these things, but I think it would be best to incorporate my thoughts into other posts, as this one is getting long and could have easily been a five part post. If you are still reading, I am sorry.

Den Haag HS

We continued on our way through Rijswijk and the residential areas of Den Haag, and finally arrived at the Den Haag HS. Every central station in every city has dedicated an enormous amount of real estate and resource to bikes. Bikes are not permitted aboard local trams and buses or regional Randstad Rail Sprinter trains. Space for bikes on intercity Nederlandse Spoorwegen (NS) trains is limited, and bikes are only allowed during non-peak hours (It costs 6.50 Euro to take a bike aboard NS trains, if you’re curious). For these reasons, the first and last mile bike parking facilities are essential. This is the multimodal strategy that many more U.S. transit agencies have begun to adopt in recent years. Most facilities have free covered parking, but some like Amsterdam and Leiden have variable rates, and premium secure bike parking as well.

Covered bike parking

More covered bike parking

Bike repair shop

Bike/ped tunnel under station

Den Haag Centraal features lots and lots of free covered bike parking, a bike repair shop, and even a bike tunnel that goes under the station! This station, like all stations (and many buildings for that matter) are really long, so rather than making pedestrians and bicyclists go all the way around them, they built an underpass. As it turns out, the Dutch are way serious about bike underpasses. I loss count of the bike-specific tunnels and underpasses I travelled through after only the first few days.

Dutch intersection design with cycle tracks

Dutch intersection design with cycle tracks

Here are a few great examples of the standard Dutch solution to signalized intersections that allocates substantial space in the right of way to bicycles and pedestrians, preserves visibility of bicyclists and pedestrians, reduces right turn (and left turn) conflicts, and does not require any additional space! Setting back yield and stop lines, and crosswalks (and crossbikes) from the intersection allows for traffic islands that create refuges for pedestrians and bicyclists, affords greater visibility at corners, and dramatically reduces the likelihood of right hook collisions. I’m unsure of the extent to which this reduces queuing capacity, although I assume it is negligible. Both the zebra striping and elephants feet clearly define the bike and pedestrian crossing paths. Separate signal phases for each mode (and direction) is of course an integral part of this design’s success. The second and third stage yield teeth are a feature that reinforce the right of way of traffic proceeding straight over traffic performing right turn movements in addition to the no right turn on red. I see these intersection designs as one of most important design applications for us in the United States. Many of our intersections can be retrofitted in a similar fashion, but the most glaring oversight with this intersection is pedestrian access. It would also be difficult for a person in a wheelchair or other mobility device to get on to the raised traffic islands due to the grade separation, so there should be a curb ramp/mountable lip that enables them to roll up and not have to wait in the cycle track. If I recall correctly, the pedestrian walk phase here was pretty short as well. There are traffic islands on two sides of the intersection enabling a two stage crossing, but they’re unsignalized which was unusual to see in the Netherlands.There is no light rail on either of these streets, but one street did have bus service. I wasn’t able to take any decent photos of this entire intersection in the Hague, but here’s an animation that illustrates the design concept well.

On Thursday, Paul Wiggenraad of TU Delft led a very informative bike tour around Delft, Rijswijk and the Hague. We witnessed a great variety of traffic calming devices and many examples of smart, practical pedestrian, bike, and transit facilities.  According to Paul, many of these designs are old, but the roads themselves are relatively new – much of them constructed only in the last four decades – as the Netherlands also experienced a proliferation of low density development and auto-centric infrastructure from the 1970’s to late 90’s. I will just tell you now that the Dutch are not big on bike lanes. They certainly exist, but are relatively few in number, as cycle tracks, bike boulevards, advisory lanes, and off-street multi-use paths (in that order) comprise the majority of the bikeway network. I plan to do a separate post about cycle tracks since they are used in a many creative ways, and represent the shape of things to come.


We left Delft heading northwest on the banks of the Delftweg towpath/canal.  The east bank of the Delftweg towpath is a multi-use path for pedestrians, bikes and scooters. The west bank is a residential/service road with standard grade-separated cycletrack. Trams to the Hague also run alongside the path for much of the way. Here is a photo of a maintenance crew either extending the bank or some form of erosion control. A neck-down for boats! We left the towpath and rode through several residential neighborhoods in Rijsjiwk. Here are a few examples of the kinds of facilities we observed.

The following photos depict a typical local residential road (“local estate road”) with advisory/”suggestion” lanes and what we in Portland would aspire to call a bike boulevard or neighborhood greenway. Here the roads feature curbside parking (6-8 ft), one travel lane (12-14 ft) , and dashed advisory/suggestion bike lanes (3-5 ft). This is the Dutch standard for low-traffic residential streets (I won’t go into the details about the criteria for their advisable dimensional spacing, but the CROW Manual does if you’re interested).The speed limit on these streets is 30 km/h or about 18 mi/h (more on speed limits later).

Advisory or Suggestion lanes

Here the road is still two-way, but the center travel lane is not quite wide enough for two cars. This requires the motorist to navigate slowly around bicycles and traffic moving in the opposite direction when space permits. This design provides flexibility for each mode, and has inherent traffic calming properties. Many of these roads incorporate red brick or asphalt mixed with red pigment (not paint or thermoplastic!) further indicating that the bicycle has priority on the street. An interesting note worth further discussion in another post: The CROW Manual recommends against coloration of only the advisory lane, reserving it specifically for “bike/cycle lane” facilities. In the Netherlands, the advisory lane is more or less applied in the same context as the bike/cycle lane (approx. <2,000 ADT), but is much, much, more prevalent. My one minor critique with this design is that it allows autos to park facing both directions on either side of the roadway. This is one of my pet peeves back home, as it makes it a bit more difficult on a bike to tell which direction cars are likely to pull out in front of you, and if you’re riding in the travel lane you have to watch both sides. Of course it’s less of a problem here because people operating autos are ever more aware of the presence of pedestrians and bicyclists.

Speed table

In this photo is a raised speed table (not to be confused with humps or lumps) at midblock. The signs indicate that the travel lane narrows, but this is more of a psychological effect that reinforces the traffic calming measure, since the profile of the travel lane itself is not really any narrower (as you can see, the travel profile is actually getting wider with a reverse bulb-out/chicane for bicycles). The speed tables forces motorists to slow for the grade change, and the islands require them to queue on either side when there is oncoming traffic, but it allows bicyclists to continue through, unimpeded, at grade. I did not observe this since it was a lower volume street, but I also suspect that if volumes were sufficient, it would have the effect of a signalized ramp meter phasing opposing auto traffic through at regular intervals and slow speeds. How cool! Another neat thing you will see at nearly every speed table are the white graduations on the ramps indicating the degree/severity of the change in grade. Longer striping indicates a gradual change whereas shorter striping indicates an abrupt or steep step.


30 km/h standard speed limit

Seven feet!

Rumble median


These photos are of a Dutch bike boulevard or “fietsstraat.” Again, the speed limit is 30 km/h and the street is two-way but its profile is barely wide enough for cars travelling in opposite directions to pass –  requiring both to slow and one to pull over. The entire street is paved in red, indicating it is a bicycle street where bicycles have full priority in the travel space. Furthermore, the presence of the stone rumble strip down the center of the travel lane necessitates reduced motor vehicle speeds when passing. Note the yield “sharks teeth” and the bike box for the (7ft!) travel lane. There are very, very, few stop signs in the Netherlands. Intersections where there are motor vehicles are either signalized or unsignalized roundabouts, but in nearly every case, the yield teeth are present to let you know what’s supposed to happen in what order. It’s beautiful and it works. Diverters mark the end of auto travel on the street, but allow bicycles to continue ahead.