International Wildlife Crossing Infrastructure Design Competition

I blogged about this contest back in December, so won't repeat all the information I gave then. However, it has now been launched in earnest. Expressions of interest are requested by 30th July, from teams of engineers, landscape architects and ecologists able to innovate in reducing wildlife-vehicle collisions in Colorado. Teams who successfully prequalify get a US$15,000 honorarium, and there's a US$40,000 prize for the eventual winner. The competition rules look fairly sensible with an multi-disciplinary jury, the winner retaining copyright on their design, but no commitment to award an actual design contract.

Past examples illustrated by the competition organisers imply a bridge as a solution, but most wildlife crossings I know of are tunnels or culverts, and I suspect they're looking for a more imaginative solution if one exists.

London Bridges: 4. Sackler Crossing

The last of four bridges I visited in London recently was also at Kew Gardens, the Sackler Crossing.

Designed by John Pawson, engineered by Buro Happold, and built by Balfour Beatty, this 70m long S-curved monument to minimalism connects footpaths across a small lake, encouraging visitors into areas of the Gardens not always well-trodden.

Opened in May 2006. the bridge is intended to give the feeling of "walking on water", and is set quite low against the lake (although not as low as I had expected from comments from other visitors). The bridge deck consists of 564 slender black granite "sleepers", supported on a hidden steel framework which in turn rests on steel piles ever 8m.

There are 990 bronze balusters, each with a curved top but no handrail. I've seen a similar design only on one or two other occasions, as most footbridges provide an upper rail, which serves both as a handrail and also to make the posts more economical by sharing loads between them. A continuous top rail is mandated in the relevant standards (e.g. BS 7818), but it's great to see it dispensed with for such pleasing effect as here.

The shape of the balusters is such that they appear essentially transparent when viewed face on, but solid when viewed at an angle. It's a charming concept, and lends the bridge a visual interest that its minimalist heritage would seem to have precluded.

An unobtrusive bridge was definitely the right response to a very tranquil setting like this, and I was particularly impressed as to how well hidden the supporting engineering was, with the impression given that the granite deck planks just float above the water.

It's a shame the water level wasn't higher when I visited, as it would definitely have looked better that way. You have to clamber down the bank of the lake to see what's really going on underneath, and no sensible visitor would do that.

The bridge certainly merits the several awards it has won, and offers a welcome reminder of the possibilities for bridge design beyond those which emphasise the structure or which feel obliged to at least display it honestly. While I enjoy minimalism in the arts, I'm not normally an admirer of it in architecture, but it makes perfect sense at the Sackler Crossing.



Further information:

• Structurae
• Google maps (the current aerial image was taken at a very early stage of construction)


• Kew Gardens information sheet [PDF]
• Walk on water at Kew

London Bridges: 3. Xstrata Treetop Walkway

The two footbridges over the Thames weren't the only bridges I looked at in London recently. I also spent a day at Kew Gardens, and this post covers one of the two very interesting pedestrian walkways to have been built there in recent years.

Opened in May 2008, the Xstrata Treetop Walkway is one half of a visitor attraction within the Gardens (the other half being the Rhizotron, a short underground tunnel with information on tree roots). Rising 18m in the air, and 200m long in total, it allows visitors to wander amidst the treetops, reaching out and touching some of the closer trees. The design is by Marks Barfield with Jane Wernick Associates. It was built by W.S. Britland & Co.

It's a considerable engineering achievement, considering the difficulties of bringing massive steelwork into the Gardens through narrow entrances and without damaging any significant vegetation. Each support had to be positioned very carefully to avoid tree roots, with the assistance of a radar survey.

I haven't been able to find out for sure how much it cost, although The Independent reports £4m (figures elsewhere are less). If true, and considering the bridge is about 1.5m wide, that works out at £13k per square metre. It's actually a bit less than that considering that each 12m span is connected by "nodes", areas with space to rest and absorb the view (one node is intended as an aerial classroom), and I suspect the £4m may include the Rhizotron too. Nonetheless, it's in the big league of landmark footbridges, even though there is no river, road or rail to cross, the structure is essentially conventional, and the spans are hardly challenging.

The money has gone into a spectacular but sadly non-functioning lift (it hasn't worked for most of the walkway's life, and is currently the subject of a legal dispute), the tree-like weathering steel towers, and the unorthodox and seemingly random geometry of the walkway trusses.

The use of weathering steel throughout is a great choice, not just because of the reduced need for future maintenance (especially valuable at height) but because it blends in well with the woodland setting. The support columns appear to have been carefully detailed so that water runs off easily, and the staircase is perforated to allow free drainage.

The walkway has a mesh floor, again eliminating drainage issues. The potential for rust to rub off on visitors' clothing has been avoided by the use of protective mesh panels and by covering the top rail in wood.

I think that in general it looks fantastic, especially from below. The tree-like columns are appropriate both in engineering and visual terms (although certainly not especially economic). As you look up through the floor mesh, you really want to get up there.

Once at walkway level, I was less happy, although this was as much due to mild vertigo as to any issues with the structure itself! The 1.3m high parapets were mostly reassuring, but the view down through the walkway grille was very disconcerting, and some of my companions found the whole experience quite difficult. The lack of reassurance meant that I concentrated more on negotiating the structure, and less on the trees that are supposed to be the real attraction.

In my own design work, I've had the question raised as to whether mesh flooring is in the spirit of the Disability Discrimination Act, as it clearly puts off visitors who suffer from more serious vertigo.

Up close, and indeed from most angles, the walkway trusses look like a jumble of metal struts thrown together randomly. If the idea is to give the impression of tree branches crossingly, it works very well.

However, I gather the pattern is in fact quite ordered, and certainly the floor bracing can be seen to be symmetrical about midspan. There are a higher number of diagonal elements close to the supports, as you would expect broadly in line with the variation of shear force.

The actual density of diagonals is said to vary in line with the Fibonacci sequence, which is philosophically appropriate as Fibonacci numbers govern many patterns in nature, including the branching of trees. It doesn't make such direct structural sense, of course, as the maximum shear force in the trusses does not follow a Fibonacci progression.

The walkway has already several awards, and in my opinion, thoroughly deserves the acclaim.

Further information:

• Structurae
• Google maps (current aerial photo predates the bridge)
• Xstrata
• Life in the trees: Marks Barfield's tree-top walkway at Kew
• Reaching for the Treetops (PDF, Kew Magazine, with many construction photographs)
• Xstrata Walkway Kew (PDF, Blueprint Magazine)
• Steelconstruction.org
• Blogs: Dezeen, Treehugger

Te Rewa Rewa bridge opens

The Te Rewa Rewa Footbridge opened on 5 June (see also opening day videos). The NZ$2.8m bridge spans 69m across the Waiwhakaiho River near New Plymouth. I've covered it here a couple of times previously, before it took on its final name.

The steel arch structure was designed by Novare Design and built by Whitaker Civil Engineering.

The photos below were mostly taken the day before opening, and are courtesy of Novare Design. Click on any image for a larger version.

I think they've done really well to carry through the original design vision into the completed structure, without any significant change. I'm not keen on the way the structural behaviour is disguised (the arch springings hide a set of internal steelwork which make the arch behave as a bowstring rather than, as it appears, a thrust arch). Nonetheless, it's visually crisp and appealing, and I like it.

London Bridges: 2. Millennium Bridge

I've held off on so many occasions from commenting on this bridge. I've visited it several times in the past, but on my most recent visit, several points occurred to me which are perhaps worth addressing here.

I will say as little as possible about the bridge's history: that can all be found online elsewhere. I hope at some stage to cover the losing competition entries here, but that's a project for another day.

For now, I'll just give a quick cost comparison with the Golden Jubilee Footbridges. The Millennium Bridge is 333m long, and 4m wide. It cost £23m in 2000, including the refurbishment carried out after it notoriously wobbled. That works out at £18k per square metre, compared to £13k for the Golden Jubilee Footbridges. The Millennium Bridge is one of the most expensive footbridges of all time (beaten only by Gateshead's Millennium Bridge at £22k, with Kurilpa Bridge and Calatrava's Sundial Bridge at a mere £10k each, although Kurilpa's figure is flattered by the inclusion of its less structurally ambitious approach spans). For those spectacular sums of money, a promoter must get something of spectacular value in return.

The designers, Arup and Foster + Partners, described the bridge concept as "a thin ribbon of steel by day ... a blade of light at night". To achieve this, the extremely low span-to-rise ratio of 62 for the cables is roughly six times shallower than in a normal suspension bridge. As a result the cables (and more impressively, the foundations) carry a horizontal force of over 2000 tonnes, under dead load alone (if you picked the structure up on its end, one set of cables could carry three Millennium Bridges with ease).

I know I'm not alone in finding that the bridge in reality fails to live up to the vision of a "thin ribbon of steel". From almost every perspective, it's dominated by the cables and their outriggers. The low profile of the cables, which pass below the deck at midspan, requires the use of very stiff steel outriggers to hold the deck and cables in the correct geometry. The apparent "wave" created by the varying angle of each outrigger adds to their visual mass, such that from several viewpoints you can't see through them at all. A conventional suspension bridge, with its lightweight hangers, would not have had this issue, with views both onto and off the bridge being far less cluttered. Of course, it's arguable whether the towers for such a structure would have had an adverse visual impact.

Another difficulty with the outriggers is the way they pass below the main deck structure. I'm not entirely clear why this has been done, other than to simplify construction. In their current form, they cradle the deck units, with the deck edge beams sitting above the outrigger crossbeams. Again, from many perspectives this makes the bridge seem very bulky, as the crossbeams add to its apparent depth (something that wouldn't be apparent on a nicely drawn elevation of the bridge). The alternative would have been for the crossbeam to be within the same depth as the edge beam, passing "through" it and giving a much lighter appearance overall. Presumably this would have added to the complexities of fabrication.

The bridge's south abutment (pictured right) has always seemed to be an unhappy way to terminate the cross-river journey, with the deck splitting in two and then doubling back below itself, to a landing area which is often heavily congested by pedestrians as a result. The obvious route off the bridge is blocked by a glass balustrade that offers a view of nothing in particular.

Riverboat and riverside path headroom restrictions mean that the bridge must arrive on the south bank at height. Given the presence of a large plaza in front of the nearby Tate Modern gallery, the obvious solution to me would have been to landscape that area and provide ramps to ground following the various pedestrian desire lines. However, I believe that site constraints may not have allowed that when the bridge was built. At least they didn't build what was shown in the original Arup / Foster / Caro competition entry (pictured on the left).

The dampers which have been added to the bridge to prevent it from wobbling undoubtedly detract further from its appearance, but it would be unfair to criticise the designers for this, as they have clearly made the best of an awkward situation. There are diagonal bracing members next to the piers, diagonal bracing below the deck, little splayed damper "legs" on the riverbank, and more. The "legs", visible in the abutment photo above right, are probably the worst addition, visually. I do wonder whether it wouldn't have been better to go the whole hog and triangulate every bay of the bridge in order to give it a consistent look, but what has been done is presumably the least possible compatible with eliminating oscillation.

To a great extent, the appearance of the bridge is a logical consequence of making certain key design decisions at an early stage, and being locked into those choices thereafter. The design concept - the blade of light - was established by Arup engineers Chris Wise and Roger Ridsdill Smith who decided they "should just make the structure from cables stretched as tight as possible between the two banks, and then walk on them". That brings to my mind a stress ribbon bridge, which the Millennium Bridge has sometimes been described as, but certainly isn't (the deck segments aren't prestressed together longitudinally, hence their potential contribution to the stiffness of the bridge isn't exploited).

The decision to use the low-rise suspension bridge to achieve this was locked in place at competition stage, and had to be made to work thereafter. The relationship between cables and deck lead directly to the very high cable forces, the unfortunate lateral frequency of vibration, and the tangle of steelwork that now defines the bridge's aesthetic, in quite marked contrast to the original concept. Some of this is a flaw in the design competition route, which generally sets the final appearance of a bridge very early, unlike "conventional" design development where different structural options can receive a more thorough evaluation (often at the cost of daring, imagination, and star quality, of course).

Further information:

• Structurae
• Wikipedia
• Google maps
• Stabilising the London Millennium Bridge (paper in Ingenia)
• Linking London: The Millennium Bridge (paper in Royal Academy of Engineering, PDF)
• The London Millennium Footbridge (paper in The Structural Engineer, PDF)
• Blade of Light: The Story of the Millennium Bridge (by Deyan Sudjic)
• Bridge: The Architecture of Connection (by Lucy Blakstad)
• 30 Bridges (by Matthew Wells)
• Bridge Builders (by Martin Pearce & Richard Jobson)
• Bridges / Puentes (by Martha Torres Arcila)

London Bridges: 1. Golden Jubilee Footbridges

I was in London at the end of May, and passed by a couple of bridges on the River Thames very briefly, en route to somewhere else. So the next couple of posts aren't intended as comprehensive reports, just as quick commentaries.


The first of the two visits was to the Golden Jubilee Footbridges, two nearly identical structures jutting out to either side of the Hungerford Railway Bridge. These were the result of a bridge design competition, won by WSP and Lifschutz Davidson in 1996 against over 40 other entrants. The aim was to replace one existing footbridge supported from the east side of the railway bridge, which was narrow and unattractive. The bridge forms one of London's key pedestrian crossings of the Thames, linking two railway stations, the concert halls of the South Bank Centre, the Thames Embankment, and various routes running north towards the city centre.

The bridges were completed in 2002 as a design-and-build project, with the detailed design by Gifford, and Costain and Norwest Holst as contractors. Some changes were made to the concept design, but its basic appearance is largely unaltered and hence Gifford can be absolved of its various visual deficiencies.

Costing £39.5m in total, each footbridge is 315m long and 4.7m wide (that works out at about £13,000 per square metre of deck). The seven spans range from 50m to 65m. The steel pylons support a reinforced concrete deck, which was erected by launching from the river bank.

Many of the bridge's peculiarities relate to the site's challenging geometry and ground conditions. One pier, the A-frame now visible on the north river bank, had to be moved to reduce the risk of setting off unexploded bombs above an underground rail tunnel. Others are positioned according to the need to align the new piers with the existing rail bridge supports, one of which still sits where Brunel's original Hungerford suspension bridge was originally supported (the brick towers were reused in the railway bridge).

I dislike the use here of a flotilla of low-height cable-stay pylons anyway, but the two odd supports (an A-frame on the north bank, and a V-shaped pylon inclining the opposite way to every other mast at the main southern pier, pictured right) mar the otherwise relatively regular layout.

The overly complex arrangement of cables and bars at this particular pier strikes me as particuarly ungainly. If you click on the picture for the larger version, you can just about make out the pylon caps, which are very clumsy, with subsidiary "angel wing" elements hung below the main pylon head in order to collect the large number of cables used.

Each bridge employs 228 separate stays, and I think all these photos make clear that's simply too many stays. The visual complexity generated seems completely unwarranted by the structural demands, and surely the bridge would have looked simpler, cleaner and more elegant with fewer stays. A key design challenge, to me, would be how the new bridges should relate visually to the existing truss railway bridge, and there seems almost an attempt to hide the existing structure, or at least heavily distract from it.

The spacing and inclination of the stays is also very odd. They cluster together at the span third points, with no stays at all near midspan, and a wider spacing towards the support positions. There is some structural rationale for this (the more vertical the stays are, the stiffer they are and hence the greater load they share - so fewer stays are required in the vertical regions, and more where they incline more steeply). However, I think it looks terrible, both because of the odd spacing, and because there are none of the more shallow stays that you would expect to see towards midspan in a conventional cable-stay bridge.

Overall, the footbridge is visually unbalanced, overly fussy, and structurally over-engineered.

Incidentally, the bridge was named for Queen Elizabeth's Golden Jubilee in 2002. The Runcorn-Widnes Bridge had previously been renamed the Silver Jubilee Bridge 25 years previously, and this makes me wonder which bridge, if any, will get the honour of being the Diamond Jubilee Bridge. I can't help but notice that a certain bridge in Sunderland is proposed to start construction in 2012 ...

Further information:

• Structurae
• Google maps
• Wikipedia
• Dynamic analysis of the Hungerford Bridge Millennium Project
• Brunel lives on: London’s new Hungerford Bridge
• Golden Jubilee Footbridges, London, UK (Structural Engineering International) [PDF]
• Dynamics of the Hungerford Millennium footbridges, UK (ICE Proceedings, fee required)
• Hungerford Bridge millennium project - London (ICE Proceedings, fee required)
• Bridge Builders
• Stress ribbon and cable-supported bridges
• Bridges / puentes

Coming soon

What with an excess of work (and play), I'm conscious this blog isn't getting the attention it deserves. Nonetheless, I visited four bridges last week in an effort to keep my fellow pontists happy, and have a number of books I'm reading and would like to cover.

So, this is what is coming soon (time permitting, and in no particular order):

• Golden Jubilee Footbridges, Hungerford, London
• Millennium Bridge, Bankside, London
• Xstrata Treetop Walkway, Kew Gardens, London
• Sackler Crossing, Kew Gardens, London
• A review of Failed Bridges, by Joachim Scheer
• A review of Thomas Bouch: The Builder of the Tay Bridge, by John Rapley
• The opening of the Te Rewa Rewa Bridge, New Zealand (see videos at vimeo.com)

I'm also considering a series of posts on books about the bridges of Britain (by Jervoise, Wood, de Maré, Wright, Richards and possibly others), and how they reflect changing cultural attitudes to pontism.

Also, if you think I'm missing out on bridge news worth reporting here, please let me know!