18 September 2008

Calgary: Calatrava or competition? Controversy!

Calgary is on the hunt for two new "iconic" footbridges to cross the Bow River. And Paula Arab at the Calgary Herald has some very interesting thoughts on how they're going about it.

The story (also reported in entertaining manner in the Calgary Sun) is that a faction in the local council would like to appoint Santiago Calatrava to design two new footbridges in Calgary, at a combined capital cost of US$40-50m, and has agreed to spend US$25m to get both designs, and the first of the two bridges built (see much more here, here and here). Robert Remington asks in the Herald why it should cost so much more than the US$4.5m Liberty Bridge in Greenville, South Carolina. The answer of course is that it shouldn't: at about a 150m span, US$10m would be more than ample for a landmark footbridge.

Paula Arab suggest it's a disgrace that Calgary wants to directly appoint Calatrava rather than holding an open competition, one reason being that a contest promotes greater public involvement and attention. She also suggest the competition route is cheaper and more likely to deliver a scheme that meets a fixed budget. Most importantly, it gives local architects and engineers the opportunity to take part.

These are all good reasons, although anyone thinking that design competitions are the best way to deliver a project within budget is clearly unfamiliar with River Wear, Stratford and numerous other examples. One problem any promoter will see with a competition is that it increases risk: you have less control over which design will be chosen (because competitors will expect a fair jury), and over who the designers are.

The latter point would seem to be at the root of Calgary's decision not to hold a competition: they don't just want a fine "iconic" bridge, they want a CalatravaTM bridge. The brand name is at least as important as the product, it would seem. Certainly there's evidence from a rash of Calatrava proposals elsewhere in North America that few people on that side of the Atlantic have heard of any other bridge designer. To them, "iconic bridge" and "Calatrava bridge" are essentially synonymous.

Arab is on less confident ground elsewhere, suggesting that bridge competitions are good because every architect and engineers learns to design them in college (they don't), and because structurally they are simpler than libraries or courthouses (the opposite is often true). This misses the reality of open bridge competitions which is that a large proportion of entries are from people who have no idea how structures work, and this is readily apparent in what they enter.

Arab also writes:

Bridges are the quintessential proof that form follows function. It's that rule that leads to the birth of such esthetically attractive structures as San Francisco's famed Golden Gate bridge, more so than attempts by architects to draw fancy, but often impractical designs.

Again, this is a naive point of view. Many of the most successful bridge designs are shaped as they are because form follows constructability rather than permanent function, as can be seen in the works of Robert Maillart, Christian Menn and countless others.

Where I would certainly agree is that a competition is preferable to direct appointment; this is on grounds of simple fairness to the market rather than that it will necessarily produce a better result. An open competition is perhaps not the best way to go, however: it wastes vast sums of money which could be more productively spent, much of it generating designs which are not very good. Calgary could instead consider an invitation-only competition, giving star names the chance to compete against both lesser-known but equally talented designers, and even a few wildcards to ensure that fresh or local talent gets its opportunity too.

15 September 2008

River Douglas shortlist available online

The seven shortlisted designs for the River Douglas Footbridge competition are now available online. This begins a public consultation with the bridge designs displayed locally until 26th September. A winner is due to be announced in mid-October.

I should declare an interest from the outset: I was a competition entrant, and didn't make the shortlist.

One thing of interest is that the competition brief may have been based on essentially a false premise. The Lancashire Remade website implies that the bridge soffit level specified during the competition may be inadequate to allow the river to remain open to navigation. They may therefore be infringing on a legal requirement, and it's unclear whether any of the shortlisted designs could therefore ever be built. I imagine the 110-0dd competitors would have liked to know this before investing their time in the scheme!

Anyway, here are the entries (designers are not identified):

Design 1 is a combination central arch span with two stressed-ribbon side spans. It's nicely detailed, with the two side-by-side structural sections cleverly done. There's one big disadvantage. The bridge will require substantial foundations within the floodplain, which the competition brief describes as peat, sand and clay tidal flat deposits: to the layman, that means sludge. Bedrock is about 20m down. Access to the site is difficult enough anyway (mostly across fields), and access to build the foundations might be a big problem. The abutment foundations outside the floodplain will also be substantial, as they will have to resist considerable forces from the stressed ribbon, which will be only partly balanced by strut members.

Design 2 is, er, a combination central arch span with two stressed-ribbon side spans. This one is more explicit about what they are (they're strictly catenary box girders in design 1), and it's less of an arch, but really it's just a variation on a theme. This one seems better engineered, but the disadvantages are exactly the same. With both the first two designs, it's unclear to what extent the bridges are "landmarks" - one of the brief's criteria was to create a legible marker to the river's crossing point, which suggests that it perhaps ought to be visible from a distance.

Design 3 is definitely the odd-one out, a puzzling steel truss-frame bridge with a short side span and a large Vierendeel-truss main span. I think this is a pretty surprising choice for the shortlist as visually it's quite out-of-place, not even matching the appearance of the old railway bridge that used to sit at this site. Again, there's a foundation in the floodplain with a high thrust component. It seems a bit arbitrary for this scheme.

Design 4 is one I found hard to like, mainly because of its odd proportions. It consists of two asymmetric cable-stays, when one would be easily enough to carry this span, and they are both unnecessarily tall and with exceptionally short back-spans. The net result (again) would be expensive foundations having to carry substantial thrust loads. Not apparent from the elevation, the cables from each mast carry a different edge of the deck, which again just seems unnecessary.

Design 5 brings with it a remarkable sense of deja vu, for yet again it's another arch-and-stress-ribbon combination. This has several interesting components, although ultimately has the same concerns as the similar designs 1 and 2. The designers knows their structural engineering history, hinting at Eladio Dieste as an inspiration with the arch made from a slender concrete shell with brick facing (Dieste would have done it all in pure brick, of course).

If I said at this point that Design 6 combines an arch and stressed ribbon again, you'd be unsurprised. And so it is. Very similar to the first entry in overall conception, with two side-by-side structures incorporating steps on the arch extrados. I find it hard to choose between the four similar designs: Design 5 has the most interesting engineering, but Design 2 looks like it may be the best engineered.

Luckily, Design 7 is something different, with a very Calatravaesque asymmetric cable-stayed bridge, albeit one where the pylons are properly back-stayed (unlike the notirous Alamillo Bridge). I quite like it: it's visually striking, nicely styled, and certainly a landmark. But the foundations will be amongst the most expensive of any of these designs, and it's not really a bridge that suits its context, a very rural location. It's perhaps too bold for such a place.

None of the bridges manage to keep their muddy footprints out of the ecologically sensitive (and constructionally difficult) floodplain, which seems odd since buildability and sensitivity to the environment were two issues raised in the brief. And I'm amazed that four out of seven are of essentially the same structural form: this will severely limit the judge's options at the next stage of the competition, and given the shortcomings of the three other bridges, I expect a stressed-ribbon option to win. I can't tell whether this represents something-in-the-air, a common solution to the site's particular problems, or is just a reflection of a keen liking for this type of bridge by one or more of the judges.

It will be interesting to see which design wins, and also interesting to see after that whether the winner ever gets built. The uncertain legal issues over the navigation are just one issue: this is also a bridge which as yet has no funding in place.

13 September 2008

Ballsbridge-Dodder winning design announced

A winner has been announced for Dublin's Ballsbridge-Dodder footbridge competition. Run by the RIAI (Ireland's equivalent of Britain's RIBA), this competition was open only to Irish architects aged under-35, so it's no surprise that there's no mention of an engineer anywhere in the announcement. The days when a man with a calculator was required to design a bridge are clearly a thing of the past.

It's hard to comment on Alan Dempsey's winning design based on the images available: it looks elegant if unspectacular, albeit in a very fetching shade of blue.
One thing about the competition that is pleasing in contrast to Britain's RIBA bridge competitions: Mr Dempsey takes home a prize of 50,000 euros, and that's even before he bills for his design fee. If only bodies in the UK would adopt such an amazingly enlightened policy!

10 September 2008

Iconic vs laconic - you decide!

Sunderland City Council have started their public consultation on the River Wear Bridge, which the Happy Pontist has covered sceptically before. There are images on show at various locations, as well as a consultation leaflet that can be downloaded. There are only two competing images being made available, with an earlier thought that they might offer an intermediate option appearing now to have bitten the dust.

The conventional option, which comes within a budget already promised to Sunderland by central government, is a five-span girder bridge:

Not even slightly exciting, and relatively risk free, depending on how much work has been done on geotechnical, hydraulic and ecological studies in the river, where there will be a number of foundations. Sunderland describe it as "tried and tested" which is clearly a dig at the alternative, untried, and untested iconic design by Spence and Techniker.

The iconic option is much as seen before, although the higher-resolution image (click on the picture below) seems to show the cable arrangement more clearly.

Sunderland allude to the likes of Jimmy Choo and Gucci by describing it as a "designer bridge" and pointedly note that the £30m extra it would cost could pay instead for 6 new schools. That £30m extra is interesting, because it implies they have a cost estimate significantly higher than the bridge's designers have suggested.

They also (wisely) draw attention to the issue of risk, stating that "the risk of construction and future maintenance costs rising is higher for a project that has unique features than for a tried and tested one". This is difficult stuff, because there will be very few members of the public who have any way of judging the real risks of an unbalanced cable-stay bridge like this. These might include dynamic behaviour problems, construction difficulties, huge foundation requirements, impact on birds, all sorts of things.

Before the decision was taken in July 2005 by Sunderland to make this option their competition winner, they recognised that it was "of a striking and unusual form ... believed to be unique". They commissioned an independent review, which concluded that "construction of the bridge is challenging but achievable" and its "dynamic performance ... is extremely complex", with designer Techniker confirming that detailed analysis and testing would be needed to review this issue. The review also concluded that the bridge concept may require amendment both to meet the budget and the possible dynamics issues. All this can be found online at Sunderland's cabinet minutes for 27 July 2005. It's not apparent to what extent the design has been further developed since then to address these points.

The Sunderland Echo is running a poll for the two designs, currently running at 92% in favour of the iconic bridge, and 8% for the low-cost option. This suggests residents are happy to pay the £263 council tax surcharge that their council says would be required. However, the Echo quotes several people disputing the cost, which they believe would be much lower. At the risk of provoking a deluge of comment from bridge-happy Wearsiders, I think the bridge's proponents generally lack the expertise to pronounce on this quite as confidently as they are doing. The bridge design competition world is littered with examples of unusual bridge designs where the real risks were never appreciated at the outset, and which ended in failure.

What still puzzles me is quite why Sunderland have chosen to pursue this public consultation. Public support for anything that allows Mackems to boast at being better than their Geordie neighbours is no surprise. But they have no sensible way of judging competing claims regarding feasibility, cost and risk, other than a natural inclination to trust the underdog (Spence) against the political establishment.

04 September 2008

River Wear bridge ready for public battle

The River Wear bridge competition story looks like it will run and run. Sunderland City Council have announced dates and details for their public consultation, which will see Techniker and Spence's competition-winning design presented alongside two less exciting alternatives. The aim is apparently not to pick which bridge to build, but to find out what the public thinks is most important when making the eventual choice: appearance, risk, cost, environmental factors etc.

Judging the public appetite for risk will be interesting: it's unlikely that however much information they are shown, they have any way of really judging the financial risks involved in a dramatic and unconventional bridge design.

The bridge proposals will be exhibited throughout Sunderland from Wednesday 10th September, and also available on a website (not up just yet - I'll link when it appears). I gather the image I featured here before isn't an "official" picture, so here's one of the official images as a taster (click on it for the high-resolution version showing the cable layout more clearly):

It's not 100% clear from this first image (more should be available next week), but it looks like the two curved masts aren't connected to each other i.e. the bridge consists of two cantilever masts with no tie-backs. If the masts are indeed connected, it's certainly a more stable design than currently appears to be the case.

Matthew Wells of Techniker is reported in this week's New Civil Engineer as saying that "it looks weird but it's a conventional structure with sculptural pylons". It would be interesting to see other examples of this supposedly conventional structural form: I struggled to find any other than Calatrava's Puente de la Mujer, which is a much smaller footbridge in Argentina.

Perhaps he meant that for Techniker it's conventional? But their website doesn't list any completed cable-stayed bridges in their portfolio, with the 12-year old Royal Victoria Dock Footbridge the closest thing. They have certainly proposed a bridge of this type elsewhere, at Bergen, also designed in collaboration with Spence, but I don't believe that was ever built.

Wells also notes that the bridge budget was whittled down (or "value engineered") to £31m three years ago. Working out at roughly £3,000 per square metre of deck, that's double the cost of a fairly simple highway bridge, and nowhere near the ballpark for such an unusual structure as this one.

Stephen Spence, the River Wear bridge's architect should in an ideal world be rejoicing this week. Not only has his "secret" design been finally made public, but the second arch of the North Shore Footbridge at Stockton-on-Tees is also due to be installed any day now. Of course, that's another bridge that (apparently) came in way above its original budget, amidst a very public falling out between Spence and Expedition Engineering. Expedition's Chris Wise was quoted at the time describing Spence's concept design as something that "would not stand up on Planet Earth".

It will be interesting to see whether there is any reaction from fellow engineers now that the River Wear design has appeared in NCE; otherwise the next stop on this trail will come with the public exhibitions next week, and hopefully the chance to find out more about this spectacular, eccentric design.

01 September 2008

Secret bridge unveiled ... and you can see why they kept it secret!

Today is a great day for followers of the wild and wacky world of the bridge design competition.

At last, the public can see Techniker and Spence Associates' design for the River Wear Crossing competition. Since they won in September 2005, their design has been kept strangely secret. Architect Stephen Spence has complained loudly about this supposed injustice, stating that "his plans had been checked by engineers and no problems had been found". The Happy Pontist commented on the kerfuffle in July.

The bridge, along with a watered-down version and more conventional option, is to go to public consultation in the next few weeks. However, images of the design have been revealed early in the Sunderland Echo and The Journal.

The iconic bridge design is simply amazing. Amazingly beautiful. And amazingly daft. You can instantly see how it would appeal to the marketing types at Sunderland Arc, the development body who promoted the original design competition. It's tall, it's curvy, it's spectacular: in particular, it would make an exciting logo for the city of Sunderland. Thrusting upwards. Dynamic. That sort of thing.

Matthew Wells, the structural designer, says that "the bridge was not a technically difficult or radically new structure". He certainly knows his stuff, as evidenced by his very readable book 30 Bridges. The local press claims it's a bit like the Erasmus Bridge in Holland, to back up the view that it's a proven design. But here's where the problems really begin. It's nothing like the Erasmus Bridge, or any other cable-stayed bridge I've ever seen. And it undoubtedly is both technically difficult and radically new.

For one thing, the twisty masts tilt towards the deck they are supposedly supporting. At Erasmus (and other expensive designs such as Calatrava's Alamillo Bridge and Sundial Bridge), they slope away from the deck, which helps to counterbalance the dead weight and traffic loads.

At Erasmus, there are cables which hold back the masts, restraining them against the enormous horizontal forces introduced by the main deck cables. In Sunderland, there are no such cables, and the tops of the masts aren't even tied together. The whole bridge is therefore held up by tall, pointly cantilevers. Which tilt towards the deck. Calling that amazing doesn't begin to capture the audacity of the concept! It's a bit like hanging the Brooklyn Bridge off the side of the Empire State Building.

The Sundial Bridge just about gets away with this sort of ambition by virtue of being a relatively lightly loaded footbridge. But the River Wear crossing will carry a multi-lane highway. To avoid vibration problems, or just the deck sagging into the river, those masts would have to be enormously strong, with huge quantities of steel and concrete pressed into service. As a bridge engineer, I personally find it very difficult to imagine.

Go back to the original competition and you'll find that the budget for the bridge was £43m, which worked out at about £4,000 per square metre of deck. In my view, there's absolutely no way a bridge of this scale and type could be built for that sum.

I look forward to reading other views. It's good that at last this design has been made public, opening an opportunity for debate, not just about the merits of this particular structure, but about the process by which it was developed (a design competition, judged in secret, and without an experienced large-span bridge engineer on the jury). I also wonder about the process going forward: judging from the Sunderland press, there are strong voices in favour of the design, not least from the leader of local government opposition, who we might assume to be relatively ignorant of structural engineering (or at least to have commented prior to actually seeing the design). It looks like local people will be asked to tell their elected leaders whether they prefer bread or circuses: low-cost versus spectacle.

No doubt the designers will continue to defend their proposal robustly. But amongst all the consideration of cost versus quality that will probably dominate discussions, I think it's important not to forget about risk. Whether it will stand up or not, for a given budget or otherwise, it's undoubtedly a very high-risk proposal. And Sunderland might be wise to look at recent examples from Glasgow and Stratford-upon-Avon for cases where high-risk competition-winning designs were dropped only after first wasting considerable sums of public money.