Sponge cities
Are green roofs and rain gardens really going to save our cites from extreme rainfall? Nikki Macdonald investigates.
About 6pm on Friday, as rain pelted down like never before, Scott Speed stared out at the lake that used to be a Northcote sports field.
Water was rushing in from both the pipes and the newly daylighted Awataha Stream, but the lake hadn’t yet overflowed.
While residents posted videos of their neighbourhood’s new swimming pool, Speed looked at the view and was pleased.
‘‘I saw it as a win,’’ says the manager of waterways planning at Auckland Council’s Healthy Waters department.
Greenslade Reserve, in the flood-prone Auckland suburb of Northcote, was just last year redeveloped as part of a $20 million project to return Awataha Stream to something closer to its natural state, before it was shoved in a pipe in the 1950s.
Friday’s impromptu lake was actually a storm detention basin doing just what it was designed to do. The existing sports field was lowered by 1 metre and underlaid with quick-draining fill.
For 364 days of the year it’s a regular sports ground. But when the skies open, it becomes an enormous soak pit preventing 12 million litres of water from flooding properties downstream.
Only opened in December, the reserve had a ‘‘baptism of deluge’’, Speed says. But by midday on Saturday, the lake was gone and the turf was almost dry to walk on.
The redevelopment is part of a global urban design movement to make our concrete jungles behave more like natural ones – creating green spaces to filter, drain and hold water that would otherwise surge over endless pavement, overwhelming stormwater networks.
As Sara Zwart, Eke Panuku Development Auckland principal regenerative design lead, puts it, it’s about ‘‘working with natural systems, not against them’’.
So is this the solution to preventing future flooding?
What is water-sensitive design, and will it save us?
You might have heard it called water-sensitive design, sponge cities or green infrastructure. They’re all variations on the same theme – trying to adapt our concrete-clad urban areas to better accommodate water.
That ranges from planted roofs, streetside trees and rain gardens, to filtration ponds, wetlands and floodable parks.
As Tim Welch, co-director of Auckland University Future Cities Research Hub, puts it ‘‘the problem is, we’ve paved over so much of the natural environment and diminished its ability to absorb rainfall’’.
Ageing stormwater systems can barely cope with existing rainfall. So with predictions that one degree of warming will cause a 20% increase in one-hour rainfall in Auckland in a 1-in50-year event, our cities need a radical rethink, Welch says.
He advocates everything from porous pavement and floodable parks to building up instead of out, increasing tree numbers and green space, and retaining wetlands as natural buffer zones.
‘‘This is kind of like a win-win situation, creating a sponge city, because we’re offering more green space, more attractive environments, more trees.’’
But sponge cities are not only – or even necessarily primarily – about flood prevention.
Stu Farrant, Morphum Environmental’s water-sensitive design lead, says green infrastructure is about improving water quality, as well as rainfall resilience.
Things like vegetated drainage ditches and rain gardens divert and filter rainwater, reducing runoff of contaminated water and debris into stormwater drains and out to sea or rivers.
Bringing piped streams back out into the open gives water more room to flow naturally, as well as creating a habitat for fish.
But everything has a limit. While rainforests soak up far more rain than car parks, even forests can flood if the rain is heavy enough. ‘‘Water-sensitive design can include more green roofs and canopy cover and street trees, but none of those will cope with a flood like Auckland had,’’ Farrant says.
Beyond a certain point, everything behaves like concrete
Nick Vigar looks frustrated. Sponge cities are great, says the Healthy Waters head of planning. They’re also largely irrelevant to Auckland’s floods.
The trouble with sponges is that once they’re saturated, they stop working. So while green infrastructure like tree soak pits or water quality ponds are good for capturing and cleaning regular rainfall, they’re only designed to cope with 25-35mm of rain in 24 hours.
The rain gauge nearest to Greenslade Reserve recorded more than 150mm in just three hours, from 5-8pm, on Friday. At those levels, all the rain gardens in the world won’t help. Especially once your soils (the sponge) are saturated, Vigar says.
‘‘The reality is, when you get an event like that, every surface behaves as if it’s impervious . . . The difference between the amount of water getting in the ground and the amount that’s falling means it might as well all be concrete.’’
What about porous pavement? While traditional road pavements are weakened by any water penetration, porous pavement is designed to absorb water.
Theuns Henning, Auckland University associate professor of civil and environmental engineering, says permeable pavement is made up of material of the same size, with gaps between, like a bag of marbles.
In New Zealand, it’s commonly used for road surfaces, to let rain soak through the porous asphalt upper layer on to the solid asphalt below, removing surface water and improving skid resistance.
But full-depth permeable pavement, where water runs through to the soil, has not been widely embraced here, Henning says. The benefits are reducing runoff and replenishing groundwater, like aquifers. But while his research has found it works, it’s up to three times dearer and not suitable for high-volume traffic.
‘‘To think that tomorrow we can convert all our road pavements into permeable pavements is simply unrealistic from a cost perspective. We won’t be able to afford it.
‘‘In the prevention of flooding, it’s not one solution that’s going to give you all the answers. You’ve got to do a little bit of everything and it will collectively make things a little bit more resilient.’’
Asked if permeable pavement would prevent severe flooding, Vigar simply shakes his head.
Niwa’s natural hazards and hydrodynamics principal scientist, Graeme Smart, says the usefulness of making surfaces more porous is also sitespecific.
Heavy clay soils can’t absorb much water. Soaking rainwater into areas with a high water table, such as parts of Christchurch, will still cause flooding, by bringing the groundwater above the surface. And increasing rain penetration into unstable hill soils could actually increase the landslide risk.
‘‘These things are helpful but they all have their limits, and when we get really extreme rainfall, everything gets saturated and these measures become much less effective.’’
Contain and carry
In extreme rainfall, only two things really help – detention basins and safe overland flows.
Which brings us back to the Awataha Stream. Not surprisingly, rainwater likes to flow where streams used to exist.
‘‘Typically, if you try and control a stream by putting it underground, the result is that you end up with flooding,’’ says
Zwart. So urban designers are now trying to understand catchments and historical streambeds, and work with them, instead of against them.
That might mean bringing streams back to the surface and creating space for flooding, as with the Awataha Stream. (They also expanded existing piping to double the surge space.)
Or retreating from land to create wetlands to protect downstream developments, as the council is considering for the 12km Puhinui Stream in South Auckland.
‘‘I think the biggest, strongest thing we can do is change our attitudes to nature, and realise that we can’t control it,’’ Zwart says. ‘‘And we need to lead our planning by understanding the natural systems that underlie it, and how we can support and work with them, rather than control and contain them.’’
The trouble with water is it doesn’t respect boundaries – either physical or political. So
the challenge is getting collaboration across multiple agencies and landowners, Zwart says.
‘‘You can’t just fix a section of a stream, or fix a wetland. You need to work with everyone around it to fix every part of the system. So I think that’s the work.’’
Vigar says while Greenslade Reserve is next level in terms of design, many of the city’s parks – including Myers Park, off Queen St – are already designed to flood, with bunds to keep water in.
But even then, a detention basin is only as good as its design brief. And, with extreme rainfall events increasing worldwide, knowing how big an event to plan for is difficult, Smart says.
‘‘If you’re going to have a system that will cope with a 500-year event, it’s going to be requiring a lot of area, a lot of investment, and it’s just a matter of balancing this against the probability of it happening.
‘‘Things are changing so rapidly that councils are scratching their heads about what sort of capacity they should be designing for.’’
The Greenslade Reserve was designed for a 1-in-100-year flood. But by the 8pm Friday peak, it was swamped and spilling over on to the road. Some nearby shops and an art gallery and library had some water damage, although it was not clear if that came from the reserve.
Kaipātiki Local Board deputy chairperson Danielle Grant says the downstream land, which previously contained 350 state houses, had been boggy for decades. The new development contains 1700 new houses.
Kainga Ora says the reserve and a similar project at Freeland Reserve in Mt Roskill appear to have held up well, with no severe impacts to Kainga Ora homes.
The revamped reserve definitely made a difference, says Grant. ‘‘It’s been a massive investment and definitely worthwhile for our community.’’
When she visited on Saturday afternoon, the lake had completely drained. ‘‘It’s amazing.’’
The last resort
What Vigar does want to talk about is overland flow paths. That’s the flood network of last resort – the places the water goes when it overwhelms everything built to carry it.
‘‘No amount of treatment infrastructure is actually going to fix what happened on Friday. What happened on Friday is about overland flow paths, and flood plains.’’
As Farrant puts it, ‘‘Once you get to a certain level, you can’t really engineer water away. You need to be accommodating it, and living with it.
‘‘To deal with a flood like Friday, it’s then all about how that water can move and pass through the city in a way that doesn’t cause loss of life and unacceptable property damage.’’
And Auckland is not alone. Farrant says if the Waimapihi Stream in Wellington’s Aro Valley breached its pipe, the flow could float cars and would have nowhere to go except through the central city.
‘‘Things could get quite dramatic quite quickly, and there’d be no pipes that could hold that.
‘‘So part of it comes with how to educate the community in those large rainfall events as to how they’re going to get from their workplace to home – or are they not going to be able to get home – and have contingency for that.’’
Farrant did a Winston Churchill Fellowship in 2018 examining how other global cities design to incorporate water.
One case study was Denmark, which uses roads as stormwater pipes. The road surface is lowered to increase the carrying capacity and the roads are closed when the skies open.
The Netherlands has floodable squares to contain downpours. Australia has rainwater tanks that automatically empty before heavy rain.
Some American cities have taxes based on the percentage of property impenetrable to water. Kuala Lumpur has a traffic tunnel that doubles as a huge stormwater pipe.
While retrofitting existing infrastructure is difficult and expensive, every redevelopment is an opportunity to rethink how to better integrate water, says Farrant.
‘‘We’ve got a long way to go.’’ Welch hopes Auckland’s floods spur real action. ‘‘Every headline is ‘Is this a wake-up call for Auckland?’ But if you look back to three weeks ago when we had Cyclone Hale, that was also a wake-up call and there are other wake-up calls. So at some point, we’re just going to have to wake up and actually do it.’’