Monday, December 10, 2007

The brilliance of freeway tunnels (part 1 of 2)

A few weeks back, Gonzalo Camacho sent me an intimidating 30-page white paper on the tunnel option for expanding the I-45N corridor using some of the newest tunnel-boring technologies from Europe and elsewhere. It took me a while to get around to reading it, but in one fell swoop it converted me from skeptic to a true believer.

The essence of what makes it so compelling is that all of the money spent is for completely new capacity, since the existing surface 45 stays right where it is. Compare that to the current alternative being proposed, which, at the end of the day after $2+ billion is spent, only adds a net of 3 new lanes of capacity between downtown and Beltway 8 (from 8 + HOV to 8 + 4 managed lanes) - and that's after 5+ years of nightmare construction (vs. disruption-free underground tunneling).

On top of that, the tunnel can also solve several problems not addressed in the current plans, by continuing through downtown to 45S, 288, and 59 - bypassing the downtown bottlenecks at the Pierce Elevated and the 59-288 junction. Talk about killing several birds with one stone.

What we're talking about here is a congestion-priced, tolled set of express through-lanes that only have a few exits at major junctions. Local traffic stays on the surface freeway, which may evolve into a more sedate parkway over time, like Memorial or Allen Parkway (although I'm more skeptical of that ever happening - given the high demand and powerful commercial interests along that freeway).

The paper details how safety and flooding are handled, as well as a myriad of additional benefits: far faster construction, a substantially longer roadway life expectancy, and air, noise, and visual pollution reduction.

Raw cost is 50% more - about $3 billion instead of $2B - but when complete lifecycle costs are considered, it only works out to about 10% more per mile. And I don't think that considers any of the value in terms of time saved by drivers in both the construction and finished stages (the surface option is expected to get re-congested relatively sooner). The financials are well detailed, and summed up in a great table on p.26.

Tip to Gonzalo: re-calculate your costs in terms of $ per lane-mile of capacity - rather than per mile. Since your option adds so much additional capacity, it has to look much better that way (and be sure to include both the surface 45 and the new tunnels in the total lane-miles capacity denominator).

Gonzalo should be commended for his passion and dedication for a cause too many - myself included - have been too quick to dismiss. You can read the paper yourself here, or browse his web site here. And if you know any of the decision-makers involved, please voice your support or pass this along. It's just too good an opportunity for Houston to ignore.

In my next post, part 2, I'll discuss my own tunnel variant ideas and the long-term potential for tunneling in Houston.

Labels: , , ,


At 12:17 AM, December 11, 2007, Anonymous Anonymous said...


Road tunnels are also fairly widely used in Sydney and Melbourne where they have not been without controversy. In 2005, there was a showdown in Sydney when a project called the Cross City Tunnel was built, then a privatization concession was offered to Hong Kong multibillionaire Li Kai-Shing to operate the tunnel. A boycott of the tunnel ensued, which led to a counter draconian move by the State government to block off nearby roads to compel motorists to take the tunnel road. The State eventually backed down, but other controversies continue to dog the project.

I won't talk about flooding or geology here, but another thing to think about is that businesses along 45 will get hit, tunnel or no tunnel. A tunnel would initially save their bacon via bypassing condemnation and sparing them construction, but it might bite them afterwards by having traffic go right by (or should I say below) them.

Infrastructure monies are scarce and getting scarcer. We must make the right decisions when spending them.


At 8:57 AM, December 11, 2007, Blogger Tory Gattis said...

Thanks, Neal. I think the fact that the tunnel will be tolled and I45 will remain free should mean that those businesses still see plenty of customers. And the current plan with four managed lanes down the middle will have very limited entrances and exits, so those people won't have much opportunity to shop those businesses either. Then there's the fact that the relieved congestion may actually draw more people down to those businesses that might otherwise avoid 45. I think it can be a win for everyone.

At 9:00 AM, December 11, 2007, Anonymous Anonymous said...

I would take issue with the notion that infrastructure monies are scarce. Capital for infrastructure is limitless. It's government money that is scarce -- or, more accurately, diverted away from infrastructure to social programs that buy votes without raising the ire of self-described enviros. The contemporary source of money for transport is the private sector. This is the result of vibrant capital markets, and the utter failure of urban politicians to make investments that truly benefit the entirety of urban populations, not just favored constituencies.

At 9:10 AM, December 11, 2007, Blogger ian said...

I think Camacho's idea is great, but I, as someone who doesn't know a thing about tunneling, do have one concern: how would this project be any different from the Big Dig? I think Camacho has said in the past that the two projects really aren't comparable, but I never heard a really good explanation of why. I was under the impression that a major source of the Big Dig cost overruns was the continual discovery and necessary capping and rerouting of unlabeled, forgotten utility lines and pipes. I'm sure we've got our fair share of junk underground too that could potentially make tunneling more expensive than predicted.

At 10:10 AM, December 11, 2007, Blogger engineering said...

Read the article that I posted at It is VERY good. I do think the tunnel technology used in the Big Dig was not the best (also a comment from an European tunnel engineer expert).

At 10:15 AM, December 11, 2007, Anonymous Anonymous said...


Your issues are noted and you are correct about large amounts of private investment being available for transportation projects. I should have stated earlier that I was referring to public monies that are available for transportation projects.

When I get home from work this evening, I'll take some time out to read Mr. Camacho's proposal.


At 1:22 PM, December 11, 2007, Anonymous Anonymous said...

Keep in mind that the planned bored tunnels under I-635 in Dallas have recently been canceled. This was after 10+ years of study and engineering including extensive core sampling along the route. Also, Dallas has ideal geological conditions for tunneling - a soft-but-strong chalky limestone.

So why was it cancelled? It was financially infeasible based on toll revenue. The cost of the tunnels was $2.15 billion, and the open trench knocked the price down to a feasible $1.22 billion.

Bottom line:
I think the $3 billion estimate cited by Camacho is too low, and there is no way the project will ever be feasible on toll revenue alone. Even with a substantial injection of funds from TxDOT, maybe $1-2 billion, I don't think it could work.

Also, the surface I-45 will need to be rebuilt anyway due to age in about 20 years (or sooner), so we will get a construction event on the surface.

At 2:02 PM, December 11, 2007, Anonymous Anonymous said...

So did I grasp this correctly? A tunnel from beltway 8 all the way past downtown?

Seriously? I mean, its certainly an option, but who would want to drive in a tunnel everyday, twice a day, for that long?

I realize subway riders do this today, but they arent exactly admiring the scenery anyways.

Problem 2. No radio stations?

Problem 3. With the limited exits, when an accident occurs in the tunnel, it will pretty much keep people locked in place until it is cleared. At least today, people can divert to the feeder road if necessary and keep things moving along.

At 2:30 PM, December 11, 2007, Anonymous Anonymous said...

One of the issues that I did not add concerning the Sydney Cross City tunnel was that the finances were problematic.

As for an I-45 related project, we need to remember that the original 1956 Interstate Highway Act had a provision which generally prohibited the collection of tolls. Over 10,000 miles of toll roads that had been previously built by State governments were incorporated into the system.

I am not sure about the current status regarding the general allowance of tolls on the Interstate system. Obviously the I-10 expansion has toll lanes in it, but it may be from a legislative perspective that there might be further enabling legislation needed to allow toll collecting on an I-45 tunnel. Still, that would not be an insurmountable task.

It would seem to me that a $3 billion price tag for a toll road would probably need some $200 million per year in toll collection in order to make service payments, start to make some paydown on debt, and to provide for maintenance. That's 100 million drivers per year (275,000 drivers per day) making a $2 toll payment per trip (not $1.25) and we saw what happened earlier this year with congestion pricing by the County Commissioners. It does make for holding one's breath.


At 4:05 PM, December 11, 2007, Blogger Michael said...


I think congestion pricing might actually work if it were done sensibly.

Congestion pricing should be regulated such that the toll operator can charge whatever minimal amount is needed to keep traffic moving at 45 mph, while still giving them a profit - the tolls would be anywhere from say $.15 (at no congestion level) up to $5 per tolling location (at high congestion level) - prices would need to be re-adjusted every 15 minutes or so.

The prices I am making up are just approximations - but the reason I oppose tollroads right now is mainly because they are not priced to market conditions. And our congestion pricing proposal was a joke - prices raised for 6 hours a day in both directions. And no corresponding forced lowering of prices to below average for times when there is no congestion - which basically makes the road useless. Who is going to pay $4 to take Westpark when you can take I-10, 59, or surface roads for free at off-peak times? Keeping an artificially high price at these times just takes the roadway away from the community.

I think the only reason this is not done already is that it is more complicated to set up and maintain initially - for instance, I would want to see the current toll conditions before I leave work on Transtar, and you would need displays on the roadways that could show variable pricing. And when there are accidents, and you are already stuck on the tollway, you should not be charged $5 per toll location. But with all electronic tolling, like Westpark, it should be possible to re-adjust tolls collected based on accidents, etc.

I wonder if any city anywhere in the world already does this. I have seen this commented on as an inefficiency of, for instance, the London system.


At 4:15 PM, December 11, 2007, Anonymous Anonymous said...

I-10 is technically not being tolled. The interstate is still free and open to all users. The tolled portion is a service for commuters operated by HCTRA.

I like the idea of the tunnel, but you can also throw my hat into the ring of never wanting to use it (just like I never want to use a bus).

Also, wasn't the Hardy Toll Road being looked at to operate as a managed lane concept? Just like US 290 will have a companion toll road. The Hardy Toll Road will go into downtown after construction starts in a year or two.

At 5:24 PM, December 11, 2007, Anonymous Anonymous said...


You may want to investigate how the Singaporeans do their toll system. I think they've been doing some form of congestion tolling for a while now.


At 8:52 AM, December 12, 2007, Anonymous Anonymous said...

There is access to the full spectrum of radio stations, cell phones and blackberries in the tunneled Hudson River crossings from NY to NJ. There is the technology to do the same with much longer subway tunnels (paid for by cell phone companies), but there is opposition because hearing other people's cell phone conversations on a train is highly annoying.

Note people do spend over an hour each way in tunnels in NY subways. You can ride the E train underground all the way from Jamica Queens to World Trade Center (i.e., through Queens clear into midtwon and down the west side to lower manhattan), and many people do just that.

At 9:11 AM, December 12, 2007, Anonymous Anonymous said...

Just because the technology exists, does not mean Houston would incorporate them. I was just noting that the paper didnt comment on the cost of such infrastructure. Im sure its not cheap.

At 10:01 AM, December 12, 2007, Anonymous Anonymous said...

On a long tunnel, you'd likely need or want to have radio accessibility so the tunnel authority can provide alerts to motorists in the tunnel (PA does this in NY by breaking into radio frequencies). In any event, my understanding is that the systems that allow am/fm radio/cell reception in the NY-NJ crossings were borne by cell phone companies, not the Port Authority. Therefore, that's one part of the capital cost equation that shouldn't be a big deal for any tunnel building project in Houston. If you use satellite radio in your car on the other hand, I'm not sure how that would work in a tunnel.

One point on congestion pricing in the real world. Places nominally without it sometimes have it another form. The East River crossings in NY are a mix of tolled crossings (Triborough, Midtown Tunnel) and non-tolled (59th Street, Williamsburgh, Manhattan, Brooklyn Bridges). People listen carefully to traffic reports and change their routes on a dime to the tolled crossings when the non-tolled crossings are backed-up. Happens all the time. It's very often the case that you can save 20 minutes getting from Queens to Manhattan by taking the midtown tunnel. If your cab driver at Laguardia ever asks you if he can use the tunnel, he's asking you whether you don't mind paying an extra $6 to try and save some time. The Midtown Tunnel is never empty despite free alternatives in its vicinity.

At 10:38 AM, December 13, 2007, Anonymous Anonymous said...

I'm not against the tunnel idea, but it seems like I-45 from Greenspoint to 610 is one of the last freeways in the country that should get a tunnel if they were to start becoming widespread. Population density along it is quite low for an urban area, almost no neighborhoods border it, the soil is less than ideal and we all know there's nothing scenic along it. There's got to be 100 freeways more qualified, including pretty much every freeway in Los Angeles, Chicago, New York, and other cities.

At 12:00 PM, December 13, 2007, Anonymous Anonymous said...


You're right about the lack of residential development along I-45 in the belt to the loop. The only other freeway less deserving would be SH-288. This seems to be a great idea for US-59 and Katy Freeway in future expansions when needed.

TxDOT will never give up existing freeway capacity. And why should they?

And a lot of this is predicated on the ASSUMPTION that freeways are horrible to look at and should be hidden. It's just as silly as the idea that carbon dioxide is a pollutant and needs to be regulated.

At 10:02 PM, December 13, 2007, Anonymous Anonymous said...

Wouldn't it be a lot cheaper to just live closer to work? Why not just buy up lots of slums for dirt cheap, level the old buildings, and build new "suburbs" right next to downtown?

At 9:32 AM, December 16, 2007, Blogger engineering said...

It is interesting to read the comments. Have to note that proposing a tunnel was as a feasible concept that was not considered when TxDOT was studying alternatives. As it turns out, basic cost analysis indicate in case of I-45 tunnel cost is comparible to the "pancake" or lateral expansion alternative.

I think some comments are "knee jerk" sort of comments. i.e. feasibility of tunnels and costs. People make a big deal of tunnel costs but they don't make any comments about a billion dollar highway interchange or the plus/minus $3 billion I-10 expansion. Maybe they don't know about it.

Somebody mentioned the Dallas tunnel project was not feasible thus it was cancelled. I think Dallas DOT had several proposal for the project - meaning consultants were willing to invest in the project because it was feasible (some of them were Europeans with experience).

Also people talk about cost like their taxes would increase in fact I think highway related taxes (gasoline) have not gone up in almost 20 years but currently multi billions are spent on highways. Also got to keep in mind Texas legislators have passed a law preventing TxDOT from converting any "free" highway lane into a toll lane - big error because highways are not free. Thus to add "new" capacity allows TxDOT to generate revenue.

On the technology aspect of communications. In technical terms we call it ITS. ITS is the future of all highways, whether surface or underground. Folks interested in highways should visit Transtar, they give free tours.

Not sure what population density has to do with highway traffic or capacity, other than the higher the density (New York, Hong Kong) the more expensive the right of way.

What I find of most interest is while other cities are willing to invest in their quality of life (when talking about highway tunnels includes many of the large city in Europe, some in Asia, and some US cities in the east coast), Houstonians are not even willing to consider it; with clarification, the district engineer at TxDOT did say that TxDOT is willing to consider it but needs the support from city elected officials.

The best thing is that organizations such as the has adopted the tunnel alternative as their only option. This is significant because the i-45 Coalition becomes the institutional memory since every six years we have a new set of city elected officials.


At 2:44 PM, December 17, 2007, Anonymous Anonymous said...

What population density has to do with is the number of people affected by the freeway in the first place. I presume at least part of the reason we want tunnels is the effect of freeways on the neighborhoods around them. If there are no neighborhoods, then a freeway isn't so bad, and why force people to drive underground?


Post a Comment

<< Home