Living roof construction.

Living roofs are self evidently a good idea. We now have a lovely lawn for sitting on in the evening sun, rather than an expanse of tiles. It will reduce the amount of water that runs off the roof, keep the building cool in summer and reduce the heat that would bounce back from a hard surface into the warming sky.



Roof Construction

The roof is constructed from I-beams. These were expensive and are engineered and imported which represents embodied energy. However, over the same span, they are said to use one sixth the timber of conventional joists. Also they are made as a kind of oriented strand board and so can be made of younger, sustainably grown trees, rather than the huge old trees required for solid beams. On top of them is a plywood, then two layers of roofing felt. On top of this, now conventional waterproof roof, we have added a butyl pond liner, drainage layer, soil and finally turf. The roof slopes at 5 degrees.



Roof layers

• Felt roofing (two layers, as normal)
• Butly pond liner (1 sheet for entire roof)
• Old lino, tarps and carpets to protect the rubber liner.
• Drainage layer - about 2cm thick layer of pea gravel
• Filter fabric. To prevent the soil washing away.
• 3 inches of soil. This has probably compressed sown to 2 inches by now.
• Turf

Edging

To prevent the soil falling off the bottom edge of the roof we made a metal cage and filled it with lovely purple slate. This works well and no debris comes out with the rain water into the gutter.

Expense and effort

If you want a grass roof that you can use, it's a lot of work and a lot of extra expense. A sedum roof that you can't go on might be less expense, but it will still cost you a great deal more than a conventional roof. Firstly you have to make the roof structure stronger, in my case expensive foot high i-beams, as specified by a structural engineer. Then there is the task of getting all that soil on the roof. In our case all the soil came from the garden, the soil that was where the building now stands, but that meant it was full of glass from the half-century when the site was used as a market garden. It took us days and days and days to sift and lift the soil. Some of it was done by volunteers, but it's a very tedious task so I didn't feel I could get volunteers to complete the work.


Above, left. Our sifting device. Right, knackered by all the sifting. Note the size of the mound!

Fear factor

There is a fear that the roof might leak which would be complex to resolve. But as long as the butyl rubber holds there is no reason to worry. And as that's buried there is no reason for it to degrade. Time will tell...

Information - lack of...

There is quite a lot of stuff about living roofs online. A lot of it is like this information, by people who have tried this way or that. Other information is about expensive systems used on corporate headquarters and well funded public buildings. This info is largely irrelevant as it involves wonderfully well designed and extremely expensive systems. Some information appears to be guarded by people who say they know how to do it and will only release it if you pay. (Never!)

So in the end, after nights and nights of research over several years I settled on the method described above.

Performance of the roof over two summers

This summer, 2008, the roof has done very well and is reasonably lush, but it has rained almost every day. Around the edges I have left trees overhanging and the grass is going much better where it doesn't het the sun all day. Last summer, the roof became dry quickly and required watering every day it didn't rain.

There are several reasons, I think, for this problem.
1. I bought cheap turf and I may have been better seeding with a drought resistant grass breed
2. Alternatively I would do without the drainage layer. Perhaps it is too effective and the grass would benefit from the soil being allowed to stay wetter longer.
3. A slope of 5 degrees AND a drainage layer may be compounding the problem. What is the optimum angle for a living roof with and without a drainage layer. I would love someone to do some research into this. If anyone IS doing research into this they are welcome to come and lift the drainage layer out on half of the roof as part of an experiment.

Water reuse

My living roof is 80 square meters and as we live in the west a huge volume of is collectable from it. At the moment I collect (a small proportion of) it and don't use it, but I'm hoping to install a 12v solar charged pump so that I can do any watering of the lawn from the stored water that has come off the roof. I have installed a straw bale pissoire and am in the process of adding a hand rinsing tap using water from the water butt. Excess water now goes down a hose to the allotments.

Things I'd change

• I might try 2.5 or 3 degrees next time. (I only settled on 5 degrees because the saw had a notch at 5 degrees so I thought it would be easier)
• Also, it would be worth delaying doing the whole lawn to do two test patches, with and without drainage layer. Does the non drained go soggy in winter?
• I'd test different grass species too.
• Also next time I would definitely thermally insulate on top of the roof as retrofitting insulation inside the roof is time consuming. If the insulation went on top of the plywood it would take hours instead of days.