Thursday 24 August 2017


I mentioned aqueducts, somewhat elliptically, back in March.  Now I'm back there because my California water-guru Juan Browne is bored with going to visit the Oroville dam and spillway repairs and has gone down-stream to Yuba City, CA, large sections of which are built below high-water on the Feather River and protected by levees. Your attention-span for hydro-engineering is likely much shorter than mine, so I'll give you an executive summary for how to make a levee water-proof. Concrete is riotously expensive not only in money but also in carbon foot-print because it employs a lot of cement:
  • Calcination by heat from calcium carbonate to lime
  • add water to make slaked lime (calcium hydroxide):
    • CaO + H2O → Ca(OH)2
  • setting as excess water evaporates
  • carbonation as atmospheric COis absorbed back to calcium carbonate
    • Ca(OH)2 + CO2 → CaCO3 + H2O
Concrete is rather effectively water-proof, but mainly because modern concrete uses Portland Cement which incorporates hard impermeable silcates in the mix with the slaked lime. That's why you see it used in bridges, drains, walls, pillars and dams. But it's also rock hard and quite brittle.  The stones of which our farm-buildings are built are stuck together with lime mortar because it is a bit hygroscopic (water-absorbing) and considerably more plastic than the sand-and-[portland]cement used to bond concrete blocks or red bricks in building work. Our rough stone walls need to have more give because the stones are many-and-varied and each will have a different coefficient of thermal expansion. As the wall heats up in the sun, the grey granite expands a little more than the red granite but the lime-mortar takes up the slack nicely so the stones don't crack.

Acequia is the name used for irrigation ditches in Andalucia. It looks like it might have something to do with aqua /water but actually the word is transliterated from the Arabic as-sāqiya = the water-bearer. When Los Reyes Catolicos finally completed the reconquista with the fall of Granada in 1492. they kept the Andalucian peasants because only they knew how the acequia system worked and, especially, how to maintain it. The acequia were contoured: snaking round the face of precipitous hills to get from source to orange grove with just the right amount of fall / flow. They knew how destructive moving water can be. Flash flood? [10cm of rain falls in some distant place and a life time's supply of firewood goes past at 30km/h].  In 1500 Spain all maintenance was main{hand}tenance. The peasants would go up every year with azadas and delicately lift out seedlings from the water course while leaving alone the vegetation beside the watercourse because their roots held the clay in place for the channel walls. Repairing with stone or concrete was a disaster because the discontinuity between dirt and concrete is an incipient failure. And concrete settles and then cracks and water causes more settlement and cracking and . . . catastrophe.

In Yuba City this summer they are repairing a 5km stretch of the levee. The money has been freed up because of the Oroville spillway fiasco has put water-damage higher up the political pecking order. One of the issues with managing the discharge from Lake Oroville was/is to keep it steady, lads. If you let out the bath upstream then the water rises up against the face of the levee saturating it. If you then turn off the tap then the water drains away to the Pacific and the water-logged levee slumps down ripping out trees and bushes whose roots hold the structure together. This is one of four ways in which levees fail [helpful if school-marmish explanation with cartoons and cross-section] and their different solutions
  • slumping or undercutting of the face
    • reinforce the face with rock or concrete too big to wash away
  • seepage through the structure
    • cut a vertical trench in the levee up to 30m deep and parallel to the river. Back-fill it, not with concrete, with a mix of bentonite clay and local dirt that will settle down and bend a little during earthquakes and settle some more. Clays tend to be water-impermeable. Concrete will, sooner or later, crack; probably inaccessibly 20m below the surface.
  • seepage under the structure
    • build a seepage berm out from the foot of the levee to slow the water down by forcing it to travel through more dirt. When it surfaces, it travels, when it travels it carries all before it.
  • overtopping
    • Sandbags!
    • Or maybe here a neat concrete wall to raise the flood level from 100 year height to 200 year height [as you fondly believe from looking backwards at the data rather than forwards through climate change extrapolation.
That was an education for me  to realise that Hans Brinker is not the only way to prevent dykes getting destroyed in a storm.

No comments:

Post a Comment