On Sundrop Farms In the Middle of Australia

[danieldwilliam]

The inestimable Andrew Ducker pointed me towards this article on Sundrop Farms. Sundrop farms use solar energy to desalinate seawater which they then put in a hydroponic greenhouse. They have a small proof of concept farm near Port Augusta in South Australia. I’ve been to Port Augusta so you don’t have to. It’s pretty arid land. They are expanding this to a 40 acre £8m facility that will produce 4  millions kg of vegetables a year. Which I think would supply fruit and vegetable for about 5,000 people.

 This is great. Fantastic and if I won the lottery I would totally have one.  The article was touting Sundrop as an end to world hunger. I was left wondering if it was.  It’s easy enough to set up one of these facilities right next to the sea. But there is a limited amout of land within a stones’ throw of the sea.  What happens if you have pump the seawater inland to the farm?

Based on data from here

Here from the NSW department of agriculture

 And

 here from a website on small scale solar arrays

 and Here from a long paper on building a very large water pipeline from Turkey to Jordan.

 I estimate the additional capital costs for putting a Sundrop type farm 10 kilometers from the coast to be about $10m US. That about doubles the cost of a Sundrop type farm from £8m for 20 acres to £16m. 

That’s going to cost about $600k in interest costs to service per year. Or about 2.4m lettuces.

The pipeline is the expensive bit.  I estimate about $0.5m per kilometre of pipe.  The cost estimates in the paper on the Turkey to Jordan pipeline have a low cost estimate of $1.5m per kilometre for 2.1m diameter pipe.  So, I’m reducing that by a third for the much smaller bore pipe needed for relatively small market garden farm using 10,000 litres per day. (Not sure if this is the current demo-farm usage or the usage planned for the 40 acre production site.)

Bit of sense check.  $500,000 per kilometre. Assume half of that is materials. $500 of materials per metre including hire of diggers, forklifts and so on. Not crazy.  So $250,000 for labour. Assume half of that is grunt pipe bashing work rather than expensive specialist control instrumentation or surveying. So $125,000 to spend on actually constructing the pipe.  A kilometre is 100 ten meter sections.  Assume two teams of 4, one building the support struts, one connecting the pipe, so 8 people on site. They cost a $1,000 each per day in total.  So that’s 15 days to build a kilometre of pipe. Or 6 ten-meter sections per day. Or one every two hours.  That seems not crazy to me.

You need two lots of pipe. One to carry the salt water in, one to carry the waste brine out.

The costs of the solar panels to provide energy to pump the water are comparatively little.  The NSW Department of Agriculture guide implies that it takes 500 kwh to shift a megalitre of water with a head of 120 metres. The 10,000 litres of water Sundrop need a day is 1% of this. So were talking 5 kwh generated over about 10 hours per day. Let’s allow for losses and power to run the controls and so and say we need about $40k of solar panels to power this system.

A caveat here that I was having some difficultly following the energy usage figures so I could well be understating this significantly. But even if I’m out by an order of magnitude then the rounded figure is still about £10m for every ten klicks inland.  

A thousand kilometre stretch of coast cultivated to a depth of 10 kilometers is about 5% of the UK’s agricultural land use.  They are using hydroponics so the yields should be higher than soil based agriculture.  I’ve seen yield improvements touted of 5 times. I’ve also seen improvements of 30 times.  Five times would take a 1000 kilometre by 10 kilometre strip to this being about equivalent to a quarter of the UK’s agricultural base. The UK meets about 60% of its food needs domestically. So we need 1.6 UK’s to grow all the food we consume, or we’d need about 6 and a bit of these coastal slots to support us in the manner to which we have become accustomed. 

The pipeline costs probably don’t scale with size particularly.  So you might well be able to pack 60,000 farms into the 10 klick strip with much more efficient pipe construction overall and not end up doubling the overall construction costs.  This might let you move a further 10 klicks inland. Maybe 30 klicks before you’ve doubled the cost of the installation again. As an aside, if you were able to move 30 klicks inland and assume 10 workers per farm and that each farm worker has .25 dependents and that each farmer needs one additional person nearby to provide support services like education for children, health, beer services and so then the population of this strip of South Australia goes from about 10,000 to about 3-4 million people.  

But the energy costs do scale.

(major caveat – these are pretty much back of the envelope figures which I did in a hurry over lunch. If you think they are wrong, you might well be right. If you are planning to base any form of anything on them I would do your own figuring out.)

Comments

[widgetfox.livejournal.com]

Where does it come on your shopping list compared with UKIP?

[danieldwilliam.livejournal.com]

It comes from a different budget from the Libertarian Alliance Fighting Fund.

Which has been cut btw as they seem very confident that they are winning without me.

. . . carry the waste brine out.

[apostle-of-eris.livejournal.com]

And at what point do you start to wreck the offshore ecology?

Re: . . . carry the waste brine out.

[danieldwilliam.livejournal.com]

That certainly requires some careful thought. It’s definitely not a good thing to be dumping highly concentrated brine back in to the ocean.

Given that any system to safely return the salt to the ocean is going to require under water pipes it is going to be costly.