Hacker News new | ask | show | jobs
by sayusyz 2515 days ago
"The current state-of-the-art in RO desalination works very well, but the cost of RO desalination driven by electricity is prohibitive,"

Why is taking salt out of sea water such a challenge?

3 comments

Desalination is cheap in day to day terms, it’s the scale of water we use that’s the issue.

Sure, we can get ~1,000 gallons of drinking water for 3$ via desalination. That works out to a monthly water bill increase of ~45$ for a family of 5. Unfortunately, crops need incredible amounts of water. So, growing corn etc with desalination would case food prices to ~quadruple.

In California we grow water thirsty crops like rice and almonds. By substituting out for less water intense agriculture such price shocks could be reduced. There's also the whole issue of producing animal feed, which is inherently less efficient than just having humans eat the feed directly.
Rice doesn't need much water. It is just easier to grow it like that.
Also transportation costs wherein desalinated water will mostly have to be pumped upstream to places where crops are grown
> Unfortunately, crops need incredible amounts of water. So, growing corn etc with desalination would case food prices to ~quadruple.

Raising crops and live stock on deserts or other marginal farmland is a huge part of this.

Is drip-based watering used at scale?
It is.

t-tape for instance is widely used commercially for irrigation of high value horticultural crops like strawberries, tomatoes, peppers etc.

also drip systems (with individual emitters, not t-tape) are used commercially in many tree crops (citrus etc).

https://rivulis.com/product/drip-tapes-drip-lines/t-tape-dri...

However for agronomic crops (like corn, wheat, soybeans, alfalfa etc), which occur in mass plantings I'm not aware of any drip applications. Certainly at minimum if drip is used for these crops, it is extremely rare and probably not practical.

At what scale?
Large commercial monocultures, rather than personal vegetable gardens
Drip irrigation is heavily relied on by commercial farms in Israel
Since seaweed can grow on salt water, can we perhaps engineer other crops to do the same?
That sounds like a recipe to completely destroy the land, for _any_ other crops, on which saltwater crops are farmed.
Even if we aren't irrigating crops directly with seawater, we may have to engineer crops to thrive on high saline irrigation anyways, as there is seawater leeching into aquifers in many agricultural centers around the world, including coastal California. As sea levels rise, we should expect this to be a larger problem and plan for it.
Assuming you would use salt water on existing land, but wouldn’t it open up for growing crops in new places, on/close to seawater?
If you want to grow crops on the sands of the beach, maybe. Otherwise you're poisoning whatever land you're using.
That’s not feasible over time. Evaporation increases the salt content of any irrigated soil over time. Directly using sea water would slowly cover your soil with a thick layer of salt.

To get around this you would need to keep the area underwater and slowly replace your water with sea water. But, at that point you’re better off just farming the ocean which already covers the majority of the earth.

Why do you think that the cost is an increase?

I pay quite a bit more than $3/kgal, and a large fraction of that goes toward funding the SFPUC’s massive infrastructure for bringing Hetch Hetchy water to the Peninsula. A desalination plant would be located on the coast or on the bay and would not require this infrastructure.

As far as I’m concerned, the problem is the capital cost of the plant and the plumbing to safely suck in saltwater and discharge brine and has essentially nothing to do with electricity.

The situation isn’t helped by the fact that, in average and wet years, demand for desalinated water would be nil, since the Hetch Hetchy infrastructure already exists.

Sewage and distribution is not included in these costs. Current systems get water for almost nothing it’s mostly distribution and sewage systems that you’re currently paying for.

As to why this is, a single pipe carrying 1,000x as much water costs no where near 1,000x much per foot. Each home might only need 1/10,000,000 the water, but it’s got to be built for peak demand not average useage. On top of this, people don’t live at sea level, so you need to pump that sea water up before you can use it.

PS: Not to mention most distribution systems leak significantly, that 3$ assumes 100% efficiency at 50% it’s more like 6$.

None of this is at all relevant to my point. SFPUC charges over $4/ccf for wholesale water. This does not include distribution costs, and this is paid regardless of whether the water is sold, leaks, or is used for firefighting. If SFPUC’s wholesale customers purchased desalinated water and managed to escape their SFPUC contracts, they would eliminate this expense.

edit: http://bawsca.org/water/rates/wholesale

That’s not exclusive price just per gallon of water.

https://sfwater.org/modules/showdocument.aspx?documentid=774...

Wholesale untreated water is 1.02$ per 1,000 gallons. Page 16: ( 0.76 per 748 Gallons delivered). Plus a fixed fee for the size of the pipe (22.67$ for a one inch pipe.)

Curious. I’m pretty sure my water provider pays the BAWSCA rate. I don’t know why it’s so different.

edit: You’re looking at rate W-24. BAWSCA users seem to pay rate W-25, which is far higher. I don’t know exactly what the difference is.

My municipality loses about 60% of its water supply due to leaks.

They buy the water from the neighboring municipality for $3/1,000 gallons, and sell it to the users of the system for $23/1,000 for the first 1,000 gallons.

Yikes! I thought I paid a lot for water, but you have me beat.
$1000 per 1000 gallons is the price in some parts of the world.

You soon realise half a gallon is plenty to drink, and you don't really need water for much else.

> As far as I’m concerned, the problem is the capital cost of the plant and the plumbing to safely suck in saltwater and discharge brine and has essentially nothing to do with electricity.

Nothing to do with electricity? What is this based on? Everything I've read indicates it's a very large cost. For example, this [1] paper puts electricity's share of costs at 44% vs 37 for fixed costs.

https://onlinelibrary.wiley.com/doi/abs/10.1111/j.1936-704X....

Seems like using our fog might be an easier idea short term: https://csumb.edu/fernandezlab/fog-collection-project
Is it maybe easier to capture the rain that falls from the sky instead?

The rain that falls from the sky is salt-free, right? It then becomes salty when it touches the ocean?

Here in the Virgin Islands, much of our personal water consumption comes via rainfall we capture from our roofs into our cisterns (big underground swimming pools). Even with that I still need to truck in water a few times a year when the rain hasn’t been sufficient.

My roof surface area to consumption ratio is such that this system works pretty well for me (big roof, small human). However with a farm the roof surface area to consumption ratio is flipped - big fields needing water and little to no roof (maybe a tool shed or garage). And you can’t increase roof surface area since that would block sun from the plants. Industrial uses also have a poor ratio. Dense usage of water in a limited size building.

Yes indeed, much easier. The problem is that the places where you need desalination tend to be the places without much rain.
There's actually water rights/legal issues to this. For example until recently it was illegal for me to do this in Colorado; the law has changed to allow ~114 gallons/year.
Separating mixed things always means putting in work because it reduces entropy. Around 3% of US energy goes just to distilling oil, for instance.