Making Things Worse

Houston’s freeways were built for Downtown-bound traffic. The freeway and tollway network is almost perfectly radial, reflecting the fact that City-led freeway planning predated the state highway department’s involvement. The initial construction occurred between 1948 and 1972, when Downtown was the only game in town. Hines’s Galleria opened in 1970; Schnitzer’s Greenway Plaza in 1973.

Every freeway approaching Downtown loses about half its capacity to a series of braided ramps connecting to one-way streets. The remaining through capacity then mixes it up in a series of three interchanges. It is a robust, resilient design, which avoids having a single point of failure. Among US cities, only Los Angeles and Kansas City possess similar layouts; the same topology was extensively used in Eastern Bloc metro systems such as Prague, Kiev, and lines 1, 2, & 3 of Moscow and St. Petersburg.

A more common arrangement is to concentrate downtown traffic on a single central spine, as is done in Seattle, Denver, Atlanta, and Miami. The proposed reconstruction of IH-45 and the Downtown Houston freeway ring deletes the Pierce Elevated in favor of a much-expanded 59 and 10, essentially replicating Atlanta.

Downtown, and the Pierce Elevated in particular, are fairly congested even outside of rush hours. This happens for three reasons.

First, the assumption that half of all traffic is bound for Downtown no longer holds. As Houston has sprawled into an ever more multi-polar urban agglomeration, the Downtown freeway ring has shifted to a more regional role. Downtown has also been the primary benefactor of most transit improvements, including three light rail lines and a billion dollars’ worth of barrier-separated HOV and bus facilities. Over 30% of Downtown workers commute on transit, while the regional average is less than 4%. As a result, the reverse commute on IH 10 and US 59 is now worse than the traditional commute.

Second, the Downtown Ring was designed before modern standards for merges were developed. The original design relied extensively on abrupt center merges, and most of these have been removed. This has made the ring safer, but less able to process crosstown traffic.

Third, most of Houston’s growth in the last 50 years has occurred west of IH 45 and SH 288. The East and Eastex Freeways, which form the northern and eastern segments of the loop, are relatively uncongested relative to the region as a whole. This tends to concentrate a plurality of crosstown traffic onto the Pierce Elevated.

To visualize this, here’s Google’s “typical” traffic for 5:30pm on Friday, culled from the averages of smartphone users running Google maps.

Typical Houston Traffic, 5:30pm, Friday

All of the macro trends are visible here. The inbound Katy and Southwest Freeways are stop-and-go as traffic queues up to process through the Downtown Ring. The inbound Gulf and South Freeways are largely at free-flow, but these two have a smaller queue where traffic waits to get onto the Downtown Ring.

Once this traffic has merged, things flow smoothly. The northbound Pierce Elevated flows well past the merge from 45/59/288, while the southbound Pierce flows well once the traffic from Houston Avenue and Allen Parkway has gotten on board. How bad is this bottleneck today? One way to answer this question is to compare the lane count approaching downtown with the lane count passing through. By neglecting the myriad ramps to “Downtown Destinations”, we can have a rough measure of the downtown ring’s ability to process crosstown traffic. At the south end, this looks like so:

Three lanes of 59 enter after the Spur leaves the mainline, while four lanes of 288 merge in. The inbound Gulf Freeway loses two lanes to the 59/288 system, but future construction will relocate the 59 connection to the existing high-level collector-distributor, freeing up the ability for three lanes to continue. We can subtract an additional lane for traffic from 59 and 288 bound for the outbound Gulf Freeway, leaving 9 total entering lanes and 7 receiving lanes.

At the north end of Downtown, the situation is similar:

Four lanes of the North Freeway mix with four lanes of the Katy Freeway. Where 10 and 45 traffic merge into 10, three lanes continue. 45 picks up more, but loses several lanes to downtown, leaving three heading into the dual-sided merge with Houston and Allen.

The current proposed schematic for the reconstruction of IH 45 envisions a redo of the Downtown Ring (large PDF). This design promises to relieve congestion by adding lanes around Downtown. The problem is, it adds more lanes coming into Downtown than it does going around it. The upcoming 288 Managed Lanes project will add still more radial capacity. As a result, the percentage of lanes available for use by crosstown traffic is actually reduced.

Here’s the north end again:

By connecting the existing HOV skyway into a new crosstown managed lane, and by converting the existing exit-only to 45 north into an optional lane for 45 north or south, the current schematic brings the number of approaching lanes on IH 10 to six. Adding managed lanes on 45 likewise brings that number to six. However, both 45 managed lanes are connected to the existing Milam Street ramp, with a one-lane connection to the mainline provided in the southbound direction only. Still more lanes are diverted into the Downtown Connector, a mirror of the 59 Spur which preserves the direct ramp to City Hall.

By the time all of these ramps have left the building, only six free lanes plus the managed lanes remain. This reduces crosstown capacity to less than 60%. The south side isn’t much better:

The current schematic eliminates the inbound 59 bottleneck at the Spur, by continuing five lanes through Midtown. However, the same bottlenecks are kept downstream; there’s a short weave with 288 traffic before two lanes go onto 45 north, merging with two lanes from the Gulf Freeway and then tapering down to three – virtually the same configuration as today.

Any transport improvement will bring with it some “induced demand.” The improvements represented by the 45 managed lanes, the 288 managed lanes, and the removal of the 59/Spur bottleneck will lead to more cars entering the Downtown Ring. If the past 20 years are a guide, an increasing number of those trips will be crosstown and reverse commuters. By substantially increasing capacity into Downtown without a commensurate increase in capacity through and around Downtown, we are creating tomorrow’s bottleneck.

It’s hard to stop a train

As has been mentioned here before, an interesting transmogrification seems to have taken place whereby the Texas Central Railway, a private group backed by major industrial concerns seeking to provide a premium service to principally business travelers – in other words, the essence of Dagny Taggart Capitalism – has nonetheless been parsed by some rural folk as an urban, liberal, Democrat sort of thing.

The initial opposition to the line was spearheaded by a young couple with a small patch of farmland in the path of a high-speed bypass of some curvy BNSF track. This is the most understandable form of NIMBYism, and in response TCR got out in front of the release of the Federal EIS to announce that they preferred the Utility Alignment, a straight shot that would parallel existing pipeline and power transmission easements.

However, by this point the opposition had coalesced in the form of HB 1889. Sponsor Will Metcalf says he thinks rail is a waste of money when we need more funding for highways, although he hasn’t stated how canceling a privately-financed railway will accomplish that goal.

HB 1889 takes an extant and arcane section of Texas law meant to give cities some amount of control over the routes of old-style interurbans, and repurposes it by expanding the scope to county governments and classifying HSR as an interurban.

I’m not a lawyer, and this isn’t legal advice, but it seems to me that Metcalf’s proposed rewrite misses the forest for the trees. Interpreted in context, the language in 131.014(d) is not a broad prohibition on interurban construction, but a minor limitation on an otherwise broadly-worded section and chapter intended to grant interurbans great leeway in route and alignment choice.

Chapter 131 was originally written to clarify that interurbans possessed all of the same powers that had been granted to steam railroads by previous case law, as well as a few novel ones – such as the ability to condemn existing street railway tracks to reach a downtown core.

Adding “county” to part (d) without striking the other portions of the section thus forces one to interpret the statute in its entirety. Which means that, to my untrained non-legal-opinion-offering eye:

—TCR can build across private property without the consent of the county; the Porters’ farm isn’t “property of [Montgomery] county,” it’s the Porters’.

—TCR can build “on or across” a TxDOT facility without the consent of the county, since the state isn’t named in the expansion of 131.014(d)

—TCR can overpass a county road without consent of the county, because 131.014(a) enumerates “across” and “over” separately, while 131.014(d) only restricts crossings “on or across” municipal property.

In order words, the bill is toothless, because Shinkansen trains don’t have grade crossings. It could conceivably be applied to railways with 100mph level crossings, such as are sometimes found in Germany and Eastern Europe, but no one is proposing to build that sort of line in Texas.

Bank Grand Central

This past weekend I had the chance to visit friends and family in Fort Worth. The Cowtown folks have recently opened a pleasantly serpentine radial tollway, the Chisholm Trail. While construction only finished recently, some of the right-of-way was acquired nearly 30 years ago. You can see it in historic aerial imagery, in the arcs of apartment parking lots and residential streets, now bounded by the new highway.

In transportation engineering this is called right-of-way banking; the act of acquiring land now so it’ll be available for some use later.

Right-of-way banking works for highways. We ought to be using it for trains, too. And in Houston at least, one spot in particular stands out: Grand Central Station.

Grand Central, 1959. c/o the Railroad and Heritage Museum, Temple, Texas

Originally constructed by the SP in 1934 (replacing two older terminals at the site), Grand Central Station was torn down in 1959 to make way for the Barbara Jordan Post Office. 55 years later, USPS no longer needs the site, but it hasn’t moved despite several years on the market. The recent announcement that downtown PO boxes will be moving to Midtown suggests the postal service is following a common Houston development strategy, “raze it and they will come.”

If the Grand Central / Post Office site passes to a private developer, it will be lost as a potential rail station site. And this would be a shame, because the Post Office site is the best spot for a rail station in Downtown Houston.

Why should trains stop at Grand Central?

It’s the closest to Downtown

Other proposed rail station sites are far outside the core business district. A ten-minute walk from Metro’s former North Intermodal Terminal (now rebranded as Burnett Transit Center) gets you to Buffalo Bayou, while a ten-minute walk from the East End light rail station at Capitol and Paige doesn’t even get you to 59. But a ten-minute walk from Grand Central gets you well into the core downtown business district.

Grand Central Station; 1/2 mile walkshed based on actual walk routes, not GIS circles

Moreover, the site is very close to the Downtown tunnel network. Existing entrances to the tunnels at the Alley Theatre and 717 Texas are only a couple of blocks away.

It’s got beefy infrastructure

You can’t just plonk down a terminal rail station on any old city block. You need room for the traffic it’ll generate: pick-ups and drop-offs, taxis, Ubers, buses. The Grand Central site has the distinct advantage of having already been a train station, so this is all designed in.

Franklin Street is partially elevated over Buffalo Bayou, to make room for seven lanes of traffic. The Smith Street extension, originally designed for six ten-foot lanes, currently carries three:

Smith Street extension has ample capacity for station traffic.

No other site in the Downtown area has this much reserve capacity for the additional vehicular trips generated by a terminal station.

Downtown evolved around it

Here’s the station in 1944, in its heyday:

Grand Central Station, 1944

And here’s what it looked like just a few years ago:

Grand Central site, 2008

Note that the existing overpasses for IH-10, IH-45, and their associated ramps are already configured to make room for railroad tracks. Note, too, that the promenade attached to UH-Downtown is designed to span multiple tracks. Very few potential station sites come with the grade separations already constructed.

What trains would use it?

Recall the reference to Chisholm Trail at the beginning of this post. When land in Fort Worth was being reserved, the route was intended to be a freeway. The era of building principal radial routes as toll roads hadn’t yet started in Texas.

Likewise, it’s not particularly important what train service utilizes the Grand Central site, so long as it’s some sort of medium-to-long distance service. What might use Grand Central?

  • Texas Central’s proposed high speed rail to Dallas, or a later extension of that line
  • An alternate high-speed line to Austin, which has been studied by TxDOT
  • Commuter/regional rail to College Station, which would link multiple A&M campuses (Biosciences, Prairie View, and Main)
  • Commuter/regional rail to Galveston, which has preserved its own historic rail station
  • Improved rail service to Louisiana, which has been a federally-authorized high speed rail corridor since the 90’s.

While the station could conceivably serve several of these services simultaneously, only one need come into existence to justify preserving the land for a terminal. And if we think on freeway/tollway timescales of 20-25 years or more, only one of these lines needs to open by 2040.

It seems preposterous to suggest that a city that’s growing as fast as Houston, in a state that’s growing as fast as Texas, won’t have so much as one commuter, regional, or intercity rail line operational by 2040.

This is especially true if you prefer the Gattis/Kotkin sprawl uber alles vision to the Crossley/GCI garden city one. For the farthest-flung, lowest-density suburbs are only serviceable with fast trains. The commute from Bridgeport or New Haven to Manhattan, at 60-80 miles, is only possible because of 90/100mph trackage installed long ago. The Utsunomiya-Tokyo run, a 70-mile haul undertaken by more than a few salarymen, takes just 54 minutes on the Tohoku Shinkansen – less time than Woodlands commuters spend on the bus.

Who ought to buy it?

The Grand Central site could pass to any number of entities, provided it was purchased by (i) a public or nonprofit agency (ii) for the express purpose of preserving land for a future terminal station. Possible owners include:

  • The City of Houston
  • The Gulf Coast Rail District
  • A dedicated 501(c)(3)

A special-purpose entity seems like the obvious choice. Such a nonprofit could pool funding from the city and county, METRO, management districts, and private philathropists. Houston uses a similar model for parks; the Houston Parks Board, a 501(c)(3), partners with other agencies to move trails and park improvements forward, while the Houston Parks and Recreation Department is a traditionally-organized municipal division tasked primarily with maintenance and operations.

Rethinking Richmond Rail, Pt. I

In Whither the Limiteds, I analyzed a few bus routes which will see substantial travel time increases under METRO’s reimagined network plan, and suggested that where surface light rail pulls buses off of an existing freeway HOV facility, rail may actually increase commute times.

METRO’s University Line, or what is popularly referred to as Rail on Richmond, seems poised to do exactly that. Here, for example, is a portion of the c. 2002 bus map, before any rail lines had opened.

Richmond, c. 2002

Here’s the same corridor, as it appears on the current map.

Richmond, c. 2014-15

And here’s the reimagined network. Note that the park and ride routes aren’t shown, but they’re still there – most are on 59, with one originating at West Bellfort and accessing Greenway Plaza via the Edloe off-ramp.

Richmond, c. 2016

Richmond Rail is unlikely to happen anytime soon. Riders in federal transpo bills prevent funds from being allocated to it, and even if that were changed METRO currently lacks the cashflow necessary to issue more debt. But the line remains on a map, and Houston history suggests lines on maps tend to eventually get built.

Who is impacted by Richmond Rail?

Analyzing the impact of Richmond Rail on the existing ridership base is tricky; one must predict the ridership distribution on the reimagined network, then look at how that will itself be altered by the rail. We can’t be exact, but we can make an attempt.

Many buses won’t be impacted by the University Line. Routes that cross it, like the 27-Shepherd, will stay the same. The park and ride routes will likely stick around, as they have in the IH-45/North Line corridor.

Current routes likely to be impacted are as follows:

9 – 1400 riders
25 – 6000 riders
53 – 3900 riders
132 – 2300 riders
163 – 9600 riders
274 – 450 riders

Of these routes, #9, #132, #163, and #274 are principally commuter routes, while #25 and #53 are long local routes which eventually make it to Wheeler Station and Downtown, respectively. The reimagined network makes the following alterations:

9 – Stays largely the same, but loses one-way running in the Gulfton area and picks up a bit of route among the thicket of inexpensive apartment complexes behind the Sharpstown Mall / PlazAmericas.
25 – Stays largely the same. Is through-routed across Midtown to Blodgett Street, providing one-seat rides to TSU and UH.
53 – Deleted. The Briar Forest/Westside segment is rerouted via Harwin and the 59 HOV Lane to become the 152.
132 – Replaced by the 151, 152, and 153.
163 – Becomes a crosstown route. The fastest route to downtown involves a transfer to the 152/153 at Fondren/Westpark. The 151, 152, and 153 combined take over the Hillcroft-Downtown commuter run.
274 – Replaced by the 151.

If Rail comes to Richmond, the following route changes are likely:

9 – Is shortlined at Bellaire Transit Center. All downtown-bound riders transfer to Richmond Rail. Peak-hour, peak-direction riders may opt to transfer to a park and ride route, in exchange for an extra fare.
25 – Is shortlined at Greenway Plaza. Alternately, is sent south along Buffalo Speedway and then through Rice Village, providing a frequent route to that area.
151/152/153 – Shortlined at Hillcroft Transit Center, providing most of the operational cost savings after LRT opens.

Travel Times

A detailed travel time analysis of the University Line would take into account curvature, acceleration profile, and driver decision-making to create a likely time-space profile. Absent this effort, a reasonable approximation can be made by seperating the line into two segments.

East of Weslayan, the line looks like every other Houston light rail line. Street running without gates is limited to 35mph by AASHTO, and attempts to minimize right-of-way acquisition will likely lead to 20mph station approach and departure zones, as found on the East, Southeast, and North lines. Likewise, 90-degree curves in Greenway Plaza and Midtown will likely be flagged at 10 or 15mph. Average speed in this segment will likely be 14-15mph, in line with other Houston LRT lines.

West of Weslayan, the U-Line operates in an exclusive right-of-way inherited from the Southern Pacific railroad. With four-quadrant gates, speed limits can be set at 55 or even 65mph, and absent gross civil engineering incompetence there should be no reason for station approaches slower than 35mph or so. Average speed in this segment will likely be 23-25mph, in line with Dallas’s DART.

Using this back-of-the-envelope calculation suggests the following travel times; changing trains @ Wheeler is assumed to incur a 3-minute transfer penalty:

Possible travel times – University Line / Richmond Rail

Certainly, there are some improvements. Hillcroft to Greenway, at 10 minutes, is substantially faster than the current #25 bus, which takes 17-18 minutes to traverse the same distance.

Other trips are a wash. The run from Timmons to Wheeler is a bit faster than the current 16-minute bus trip, but riders changing from the 25 will see that improvement eaten up by the transfer penalty. 9-Gulfton travel times increase by 5-8 minutes, but this is balanced out by increased reliability; the 9 routinely gets stuck in 59 traffic, since it enters too late to use the HOV lane.

It’s on the expresses from Hillcroft to Wheeler & Downtown where the pain will be felt. Current scheduled Wheeler-Hillcroft times on the 132 are 15 minutes in the peak, 13 minutes in the off-peak. Downtown-Hillcroft times on the 163 are 17 minutes peak, 15 minutes off-peak.

For the TMC crowd, Richmond Rail will add 8-10 minutes to the Hillcroft-Wheeler Station leg. For Downtown commuters, Richmond Rail will add 14-16 minutes in each direction. The present #163 is Metro’s highest performing route*. When the existing 53, 132, 163 and 274 are consolidated into the 151/152/153 series, it is likely that total ridership will be in the range of 12,000-16,000 daily boardings. Increased travel times due to rail thus represent a loss of three to four thousand person-hours each day.

Other Alternatives

METRO previously considered a strict Westpark alignment, which was supported by the same suburban politicians who have blocked federal funding for Richmond Rail. This option was rejected because it would miss Greenway Plaza; the stop would be on the wrong side of the freeway.

A monorail was proposed in the late 80’s, under the Whitmire administration. Opposition from residents of Afton Oaks was a major factor in the proposal’s defeat, which is why the current Richmond Rail proposal jogs south on Timmons.

Reconstructing the US 59 HOV lane as a bi-directional facility and adopting a truly dynamic pricing based on a 45mph average speed could provide rail-like travel times and reliability to the proposed 151/152/153 routes. Service to Greenway Plaza could be provided with a median express bus stop, or what Seattle brands as a Freeway Station.

The no-build alternative, mandatory in any environmental study, would ensure no one sees increased travel times… and no one sees improvements, either.

In Summary

Richmond Rail’s substantial travel time increase between Hillcroft and Downtown derives from two factors. First, the Richmond Rail alignment is notably slower than the existing HOV lane; this adds about ten minutes. Second, forcing riders to transfer at Wheeler is slower than operating direct to Downtown via Smith/Louisiana Streets; this adds about six minutes.

It is unlikely that any service which requires a transfer at Wheeler can match the current #163. However, opening a second LRT line parallel to Main Street would be wasteful, and would be a mistake given commute patterns in a multi-polar city like Houston. That, then, leaves three options:

(i) Build Richmond Rail as currently envisioned, with 15-minute commute time increases for residents of Sharpstown, Chinatown, and Fondren Southwest.

(ii) Abandon the idea of rail, and operate multiple concurrent bus services with some going direct to Downtown. Focus on improving the HOV system.

(iii) Attempt to design a Hillcroft-Wheeler rail alignment which is sufficiently rapid to get total travel times (including transfers) within about 3 minutes of the current schedule.



*If the 81 Westheimer-Sharpstown and 82 Westheimer-West Oaks are counted as a single route, the 163-Fondren falls to second place. On Saturdays, ridership on the 2-Bellaire exceeds the 163-Fondren.

Whither the Limiteds

Tonight, let us bid adieu to the 44, the 56, and the 79. All of these buses have long provided local service through North Houston, in the “second ring” between the Loop and the Beltway, before running express to Downtown through one of the nation’s first HOV lanes.

Not for much longer. With the formal adoption of Metro Reimagining, these routes will have their nonstop legs cut, instead being combined with other local routes. For 79 riders, the pain will be weakest; the scheduled travel time on the rail is only 4 minutes longer than the bus, and the old 79 was notoriously late.

For 56 riders, it’s worse. The limited makes it from Airline @ 45 to Downtown in just 17 minutes*, courtesy of the Independence Heights HOV ramp. But with the route diverted to Northline transit center, riders get to experience a transfer penalty, additional rail travel time, and additional bus travel time. All told, that’s an 18-minute hit**. That’s over a half-hour lost every single day, more than a weeks’ worth of waking hours every year.

As for the 44, time will tell. Though originally planned to end at Northline, the 44 is now proposed to take over the Main Street local route, formerly route 9; no rail transfer will be possible. The 9’s scheduled time from Tidwell/Montgomery to Downtown is 43-45 minutes, compared with 30-36 on the Acres Homes Limited. But with no transfer penalties or diversions, this may prove to be the least-worst option.

The Northline rail extension currently carries a bit over 4000 people, whereas the 56 and 44 buses combined carry 7300 daily boardings. When the dust settles, we might end up with a rail line that has increased average travel times.

How Did This Happen?

To understand how the commute-lengthening Northline extension came to be, it helps to look at the development of the rest of the system, including the original, spectacularly successful Red Line.

Metro’s Early Rail Success

The principal operational advantage of fixed-guideway transit is the ability to replace many small vehicles with a few larger ones. Rail costs more to operate per vehicle-hour, but that vehicle can carry a lot more passengers.

Prior to METRORail, the TMC-Downtown corridor looked like so:

Texas Medical Center, from a January 2002 system map

Of the 8 routes in the image, all but #1 and #34 were deleted/curtailed on rail opening. As of 2000-2002, they had the following frequencies:

002 – 10 peak-hour trips, 4 off-peak trips
004 – 3 peak-hour trips, 3 off-peak trips
008 – 4 peak-hour trips, 4 off-peak trips
015 – 6 peak-hour trips, 4 off-peak trips
170 – 5 peak-hour trips
291 – 8 peak-hour trips

This schedule (36 peak/15 off-peak) was replaced by 10 hourly trains, a clear efficiency gain. Moreover, the San Jacinto/Fannin alignment is largely 35mph trackage; scheduled speeds for METRORail in 2015 are 3-5 minutes faster than the equivalent 2000-2002 bus routes. And while the #1-Hospital bus route was preserved for institutional reasons, Reimagining will remove its four hourly trips from the schedule. Peak-hour trip counts in the Downtown-TMC corridor will have been reduced from 40 to 10, a 75% reduction.

You just can’t argue with that.

Heading North… Slowly

Now we return to the Northline extension. Here’s the Downtown-Northline corridor, again from the same 2002 system map:

The IH-45 corridor, from a January 2002 system map

The implication is clear; Metro should run trains along Interstate 45 (or a similarly fast parallel alignment), removing a majority of the myriad bus trips that operate along the freeway.

Instead, the rails got sent up Fulton, replacing the #15 bus route. The Northline LRT extension has provided some support for the concept of rail bias, nearly tripling ridership over the #15. Problem is, that’s still only 4200 daily riders, or 815 riders per mile. This is well below the systemwide average for New Starts LRT lines, but does have the distinction of beating out Hampton Roads, VA.

But why have travel times increased? To understand that, let’s take a ride along the line and find out. Operating speed through the Burnett area is 15/20mph. Rolling through the Near Northside, we pick up speed, only to come across this 10mph corner (4:05 in the video):

Main @ Boundary, 10mph curve

Notably, most of the land on the inside arc here is vacant. A 20mph curve*** could have been constructed for a bit of grass and this bungalow.

A couple blocks later, we come across another 10mph zone (5:50). This tight-radius reverse curve avoided a McDonald’s, which was constructed in 2005:

Boundary @ Fulton, 10mph reverse curve

A few more blocks and one finds a 15mph left-hander (7:55). Here the alignment shifts from one side of the right-of-way to the other, resulting in track curvature that is actually tighter than both of the adjacent travel lanes:

Fulton between Hays and Irvington, 20mph curve

What do we get out of this serpentine alignment? Principally, better neighborhood service – and better potential for redevelopment. Some stops, such as Cavalcade, are better-positioned to generate infill than a similar station along the freeway would. Other stops, such as Lindale Park, likely wouldn’t exist on an IH-45 alignment.

East End Local

The East and Southeast lines will not replace any freeway flyers. The downside is that they won’t be able to experience the ridership boost that Northline will get when the 56 is cut back. The upside is that nearly everyone who rides them will see improved travel times.

The East Line will serve as a drop-in replacement for the 50-Harrisburg bus. It’ll start with an installed base of 3000-3500 transfer riders, who will have a forced transfer at Magnolia. Transfer penalties will be minimized by takt scheduling; the shortened Harrisburg route will run on 12/18 minute headways, and meet every train.

The Southeast Line is even more solid. Initial transfer ridership will come from the former 5-Southmore, 30-Cullen, 77-MLK, and the outer tails of the 40-Telephone; all of these will be reconfigured as crosstown routes. Additional riders may be poached from the current 52-Scott, which remains largely intact (but still loses to rail travel times).

Finally, the campuses of UH and TSU provide the option for some induced demand, as the Southeast Line provides a one-seat train ride from the dorms to Downtown. This may also displace some trips that are presently made by taxi, Uber, or bumming rides with friends.

Ups and Downs

Ultimately, METRO is an executive agency in a strong mayor town. METRO has pursued a policy of slower, neighborhood-oriented rail lines which do a good job of serving inner-core trips.

Where these lines replace local bus service, transit service is generally improved. The usual cost/benefit arguments can be trotted out to argue for or against the rail line, but from the riders’ perspective, things are good.

Where these lines replace limited or express bus service, transit service is worsened. Slow, surface LRT provides inferior service to riders from the second ring and beyond who are trying to get downtown.

For the time being, the Northline extension stands alone in paralleling an existing freeway HOV facility, which means the negative effects are contained. However, future LRT construction in existing HOV corridors may have similar deleterious effects. Before more rails are laid, we ought to consider the impact on existing transit riders in the second ring and beyond. Urban redevelopment is a worthwhile goal, but not at the expense of reduced service quality.


*Some rush-hour trips are scheduled to take as long as 19 minutes; midday and evening trips are scheduled at 15. 17 seems an appropriate weighted average.

**Additional route distance from the Airline timepoint is 4200′, which is a bit over 3 minutes’ travel time at a 15mph average speed. Scheduled LRT time to Pierce is ~9 minutes longer than the current bus, and average transfer wait time is 6 minutes. Walk time from the Northline Transit Center bus island to the LRT platform is neglected, since some bus routes may relocate to curbside post-Reimagining.

***For the purposes of this and future posts, any light rail curve speed expressed without qualifications will be based on 4 inches total superelevation. This is conservative; most New Starts agencies allow 5 or 6, while FRA regulations for mainline rail (which have been lambasted by many a transit blogger) allow 7.

At issue is spiral length. Bi-articulated LRVs don’t like short super transitions, so many operators hew to a 1:500 relative grade. Design criteria supplied to me by various New Starts agencies also reveal a preference for limiting unbalanced superelevation to a portion of balanced/actual; this is very similar to the AASHTO “Method 5” tables which most state highway departments incorporate by reference.

Cultural Undertones in HSR Opposition

Kuff reports that opposition to the Dallas-Houston Shinkansen has coalesced at And I have to say, I’m intrigued by what they’ve done with the logotype.

The standard TCR roundel consists of a stylized white Texas with a red line representing the railway. The state is represented by a neutral color, while the rail line line is red, representing speed and vitality. The Symbolism Wiki informs us that dark red is associated with vigor, leadership, and willpower; all appropriate connotations for a 220mph train. A blue background ties the logo into the Texas state flag.

The standard TCR roundel.

By contrast, the opposition renders Texas in the same dark red, and uses dark blue to represent the rail line. The color palette remains identical, and one can be forgiven for not noticing the difference. But the symbolism is clear: Texas is a red state, and the rail line is a blue interloper.

No Texas Central’s version.

This ought not to be the case; HSR’s core ridership base is the professional classes, and a privately-financed train could be seen as by and for the capitalists. But tribal boundaries don’t always hew precisely to economic ideology.

I’ve previously argued that cultural bias against the other tribe’s preferred transport is often unconscious. But here, it’s pretty overt.

Forgot About Delin

When Seattle’s Sound Transit looked at extending Tacoma’s LINK light rail, they nixed a Tacoma Mall alignment early on, because the grades on 25th and Pacific were too steep.

Thing is, the streetcar companies figured this out 120 years ago, which is why they carved out a diagonal street to provide a low-grade route up the hill to South Tacoma. The Lincoln District formed where it did because that was where all the car lines converged to Downtown.

Similar bits and pieces exist throughout the city. Baker Street was carved out as a switchback to allow 6th Avenue cars to make it down the hill to St. Helens. 45th at Orchard was realigned for streetcars, as was Tacoma Avenue; note the still-visible trackway.

All of which is to say, if you’re considering a new rail alignment, you ought to find an old map or three and familiarize yourself with what the original system looked like. Your problems may have already been solved.

Sound Transit’s commitment to “sub area equity” ensures that as Seattle’s rail is extended to Bellevue and beyond, Tacoma will continue to see rail expansion as well. And if there is to be a train to the mall, why not take the direct route?

It’s worth considering.


Distance-Normalized Ridership

One of the most frequent criticisms of Dallas’s light rail implementation is that it carries comparatively few riders for the amount of track that has been built. This is then followed by either criticism of suburban-oriented transit generally, or by praise for Houston’s more urban-centric system.

Recently, it occurred to me to look up the ridership per mile of the Shinkansen network. Using annual figures (since the Shinkansen has a different weekly distribution than commuter-oriented LRT), the bullet train carries 340 million people over 1484 miles of track. DART carries 28.5 million people over 85 miles of track.

That gives the Shinkansen an annual ridership per route mile of 229,101, while Dallas clocks in at 336,096.

Is Dallas’s light rail 50% more effective than the bullet train? It is if you use the same math Houstonians have been using to argue why our system is better than the big D’s.

Clearly, a new metric is needed. I propose that we evenly weight both speed and ridership density. Density, because a system that serves denser-developed areas is more effective than one that serves a few desolate park and ride lots. Speed, because a system that gets you there faster is better than one that’s slow.

Distance-Normalized Ridership takes passenger volume, divides by system length, and then multiplies by space-mean speed. If all other factors are held constant, this number can be raised by an increase in ridership, an increase in average speed, or a decrease in total system length. This also captures one of the most basic tradeoffs in transit planning; do we have more stops, and serve a greater area? Or do we have fewer stations, to provide a faster trip? Finally, Distance-Normalized Ridership also has the advantage of being constant across Metric and Imperial units, because the distance factors cancel each other out. I propose that Distance-Normalized Ridership be expressed in units of riders per hour squared.

To show how this works, let’s look at Houston and Japan.

The combined ridership of the Tokaido, Sanyo, and Kyushu Shinkansen lines between Tokyo and Kagoshima is 219,513,000 passengers per year. This line stretches across 1326 kilometers and takes approximately 6.5 hours to traverse, for an average speed of 204 kilometers per hour.

Dividing 219,513,000 into 1326 yields 165,545 annual riders per kilometer. Further division yields 18.95 hourly riders per km. Multiplying this figure by 204 km/h yields 3866 riders per hour squared, which sounds like a reasonable number. The statistics for the Tokaido Shinkansen alone are 6337 riders/hr2, reflecting the much higher ridership density on that segment.

As of this writing, Houston MetroRail carries 45,751 riders on an average weekday, plus 18,656 on Saturday and 14,494 on Sunday. This adds up to 261,905 every week, or 1,559 per hour. Dividing 1,559 into the system’s 12.8 mile length yields 122 hourly riders per mile, and multiplying by its average speed of just over 15mph yields 1871 riders/hr2.

These numbers make intuitive sense. Japan’s Shinkansen is twice as effective as MetroRail, and the core Tokaido Shinkansen is twice as effective as the network as a whole. The figures are small enough to not be unwieldy (e.g. we’re not talking millions), yet large enough that we can get a reasonable degree of accuracy without using a decimal point.

So what about Dallas? I’m not going to compute them for this post. First, because I’m more interested in proposing a new metric than in producing yet another 713-214 comparison. Second, because I’m unsure of how to weight Dallas’s various lines. DART’s slowest segment also has the most interlining, and how you weight that is a major determinant in systemwide average speed.

What I’d really like is for some outside party – say, Greater Greater Washington’s Matt Johnson – to tally up the distance-normalized ridership for Dallas, Washington, and a number of other systems. I have a feeling WMATA would score fairly decently, given that system’s relatively high speeds.

Infill and Alleyways

As mentioned in the Prologue, this blog takes “Houston basically gets it right” as a default starting point for analysis. To that end, any post suggesting we do things different will first look at the status quo and its pros and cons.

So let’s talk about infill development – specifically, townhomes and midrise.

The Status Quo

There is a long-established disconnect in Houston street design. The planning department, which approves plats, requires a 32′ street section, while the engineering department, which approves streets, only requires 27′ (28′ back-to-back). Everyone submits their plats without curblines, and once they’re approved they build the 27′ street. This gives fire trucks a 14′-15′ drive aisle in which to navigate.

As a backup, a separate portion of the city code allows the Fire Department to restrict parking on one side of any street less than 32′, or both sides of any street less than 26′. If the Chief doesn’t like what the engineers approved, he can order the signs himself.

What everyone agrees on is that a residential street right-of-way needs to be 50′ wide. At full buildout, that looks like this:

This is the street that we build everywhere from Montrose to the Katy Prairie.

The Issues

A 50-foot right-of-way is a fine width for single-family suburban development. Many a New Urbanist has fought tooth and nail to allow one-lane, 26-28′ residential streets in juridictions that required more. Some cities in California and the Southwest require streets as wide as 44′. Nationally speaking, our traditional 27′ is pretty darn good.

But that same street is overkill when faced with infill townhomes on 25′ lots. We know this, because nearly every townhome development in Houston is approved with narrow private alleyways. They’re built in Westchase, off Washington, in Oak Forest and The Heights and EaDo, in the FirstSecond, Third, Fourth, and Fifth Wards.

Clearly our developers think this is acceptable (they’re building it), and our planners and engineers think it’s acceptable (they’re approving it). And in the rare case where infill townhomes are built with a full 50′ section, they look out of scale.

The Alternative

Why not allow narrower public streets? A prototype already exists in the historic brick streets of the Fourth Ward. These streets are so beloved by Houstonians, their removal has (as of this writing) been successfully stopped by court order.

A continuous grid of narrow streets would allow for incremental redevelopment in a way that the current gated warrens of easements do not. Someone 30 or 50 years hence could buy up three or four townhomes and build a midrise apartment block, continuing the gradual, organic densification that has made inner-ring Houston so livable already.

A modern cross-section based on the Fourth Ward’s one-way, one-row-of-parking arrangement can preserve the same 14-15′ drive aisle while giving pedestrians a bit more elbow room.

An ideal street for infill development.

The narrower visual field of a 35′ right-of-way slows traffic and helps pedestrians to take ownership of the space. It also reduces land and pavement costs for developers pursuing infill development. This helps encourage the development of a fine-grained, walkable street grid in areas where it’s currently lacking.

Consider the Alleys

Houston currently requires all alleys to be 20 feet of solid concrete. But if we are to allow streets to be 35′, 20′ for an alleyway becomes overkill.

Narrower examples exist here and elsewhere. The unpaved alleys in Montrose are mostly 12′, including the ones behind Numbers and other businesses on Lower Westheimer. Many of the alleys in Fort Worth are 12′. And the City of Olympia‘s standard alley section explicitly specifies a 12’ ROW with two 3-foot tire paths spaced 3 feet apart.

I’m not completely sold on the long-term performance of tire paths. But other designs, like waffle block with a solid central gutter, could convey Houston’s design storm while allowing lesser rain events to percolate into the soil.

Notes on Peaky Transit

Alon Levy and I are having a pleasant discussion about the pros and cons of American-style commuter rail systems.

One characteristic of most American commuter rail networks is that they are “peaky”; that is, there is a substantial difference in service frequency between peak and off-peak hours. Alon sees peakiness as prima facie evidence of underutilization, something to be avoided. Let’s explore some of the reasons why “peaky” transit occurs.

Low density leads to peaky transit

The Main Line to Paoli used to have 15-minute, rapid-transit type service up through the 1970s. The infrastructure is still there to support it, but off-peak service today is every 30-40 minutes. Two things happened there. First, Center City declined in regional importance even as it grew in absolute terms, as places like King of Prussia and Great Valley siphoned off commuters. Second, a large amount of low-density residential development occurred which was not near Philly’s historic rail lines. A 1950’s student commuting to Bryn Mawr/Villanova/Haverford/St Joe’s was likely to live near a streetcar or regional rail line, enabling an all-transit trip. A present-day student is not.

In matrix form, the transit-density relationship looks like this:

Causality runs in all directions. Just as a decline in (relative) density moved the Main Line away from rapid transit, so has the gentrification of Chicago’s North Side moved that area toward more frequent service. For a long time, the Ravenswood L terminated at Belmont on nights and weekends, but in 2000 it was extended into the Loop at all times. Platforms were lengthened from 2009, allowing 8-car trains. More service was added this year, and a future project will add a flying junction to send northbound Ravenswood trains over the top of the Evanston/Howard tracks.

Likewise, low density areas rarely warrant rapid transit style service. In many cities, this leads authorities to propose extensions using different technology, with a forced transfer. Examples include eBart, the Denton A-train, and diesel trains that end at Fannin South.

Seating riders leads to peaky transit

Un-peaky, rapid transit-style services typically see a wide fluctuation in car occupancy. A DC Metro train at late morning may have only 30% of the seats taken, while one operating at the peak will be crammed to crush capacity. The same holds true for the NYC Subway, BART, and the Tokyo suburban network.

Alon lists a 2:1 ratio of peak-hour services to midday services as his ideal upper bound for rail. If we assume peak hour load factors are 2.25 seated capacity (crush), while off-peak load factors are 0.30, Alon’s 2:1 service peak is carrying a 15:1 ridership peak.

Now imagine a system where commutes are very long – say, over 60 minutes. Riders with such commutes immensely value the ability to remain seated, as mentioned by commenters here and here. If we design the service so that peak loading is 1.0 and off-peak is 0.3, our 15:1 ridership peak requires a 4.5:1 service peak.

Let’s also suppose this train consists of loco-hauled diesel trainsets, which are fuel hogs. On an unelectrified line, there is a substantial financial incentive not to run trains at 30% capacity simply to maintain an arbitrary frequency target. You might instead consider a service cutback at 75% peak occupancy. If we keep the 15:1 ridership distribution of the city metro system, but change peak and off-peak load factors to 1.0 and 0.75, respectively, the service ratio is 12:1. This is a system with 5-minute peak hour trains dropping to hourly midday (or 10/120), which is well into Chicago Metra territory.

The desire to provide seats, and to not run mostly-empty loco-hauled trains, is why brand-new commuter rail systems are more likely to emulate Metra’s peaky service than Paris’s RER. SLC’s Ogden-to-Provo Frontrunner – which largely operates over its own track – drops to hourly at midday. New Mexico’s Rail Runner has no midday service on weekdays, despite zero freight interference. (BNSF actually wants to abandon the Raton Pass line now that they’ve double-tracked Abo Canyon).

Transit bloggers like to imagine that if you provide frequent all-day service to the ‘burbs, you will open up a substantial new market of inter-suburban trips. However this ignores the fact that most of those people will just drive. And while institutional inertia can explain service provision on Metra or the LIRR, it doesn’t explain why New Starts systems, unencumbered by freight or legacy bureaucracy, nonetheless trend towards the same operating patterns.

Peaky transit is self-reinforcing

One of the characteristics which subways and freeways share is that they tend to prod people toward alternate work schedules. The early birds get up even earlier to “beat” the traffic; the late-risers negotiate a delayed start and go home when the worst of the afternoon rush has passed.

Beyond the unpleasantness of a packed subway car or a jammed highway, undifferentiated services are also faster in the off-peak. Dwell times are shorter, average speeds are higher. By contrast, peaky services actually get faster during the peak, because the ridership spike leads operators to provide different service patterns.

Let’s take Metra’s UP-West line as an example. Most of the morning service follows a “two-step” pattern in which trains originating from the outer ‘burbs run express from Elmhurst or Glen Ellyn and local trains slot in behind them. But with the 8:05am departure from Elburn, trains make all stops. What was formerly a pleasant 41-minute ride from Wheaton now takes a full 56 minutes.

UP-North has an even smaller express window. The first proper express out of Waukegan doesn’t arrive at Ogilvie until 8:19am; the last arrives at 9:15. The peakiest express in the US must surely be on SEPTA’s Paoli/Thorndale line, the artist formerly known as the R5.  The Great Valley Flyer, trains 9526/9561, runs once a day, pulling into Suburban at 7:51 and departing at 5:08. If you can swing it, this train cuts a full 26 minutes off the all-stops local.

Whom shall we serve?

Since rapid transit-level service can only exist in high-density corridors, we’re left with a few options.

(i) Allow the city to grow as it will, and only serve the high-density corridors. Houston’s light rail has taken this approach, although it is mitigated by the existence of an extremely peaky commuter bus network. This helps promote a core of hip, walkable neighborhoods, but leads to a lot of auto-based reverse commuting.

(ii) Attempt to restrict low-density development, nudging everyone towards high-density living. Portland and Vancouver have followed this approach. While this “works” from a ridership perspective, it enrages the red tribe, leading to incoherent George Will columns and pushing growth into more accommodating jurisdictions.

(iii) Attempt to serve everybody with an overlapping set of peaky and non-peaky services. Salt Lake (Trax/Frontunner), Seattle (Link/Sounder), and Los Angeles (Metro/Metrolink) hew to this format. Vancouver gestures in this direction with the West Coast Express, although they really ought to have some sort of Abbotsford/Langley service.

(iv) Build a fast LRT/S-Bahn service that hits 80% of the metro but leaves out the outer-outer ring. Dallas is there; the trains get you to Plano and Rowlett, but there’s nothing in Frisco or McKinney. Denver will be soon; the LRT hits most southern suburbs, but permanently leaves out Castle Rock and Colorado Springs.

All of these options have pros and cons. I have a strong personal preference against (ii), since I believe people should be able to live how they like. Houston has made (i) work, but this is largely because we built out the bus network before we started on rail. If Houston had instead tried to build out an “express streetcar” LRT and outer-suburban HOV at the same time, there would’ve been endless debates of the “why are we spending money on buses to the suburbs when we could promote walkable development” sort.

(iii) and (iv), then, seem like the best options for most North American cities. That doesn’t mean there isn’t room for improvement on legacy systems. Additional Metra-CTA connectivity would be good, although I think the Chicago transit geeks are needlessly dismissive of the transfer opportunities that already exist. And the LIRR has some downright questionable operating patterns which Alon has devoted many pages toward improving.

But if we’re going to allow people to live in low-density suburbia, we ought to make peace with peaky transit. Whether that takes the form of the Great Valley Flyer or the West Bellfort conveyer belt matters not.