Hacker News new | ask | show | jobs
It's 2024 and drought is optional (asteriskmag.com)
50 points by vwoolf 758 days ago
10 comments

This article is quite long-winded, but to summarize, they are proposing a desalination scheme assuming cheap solar energy.

Given that California electricity is pretty expensive even though we have lots of solar energy, how can that be done? Why not sell it to the grid?

It might help to know that the author of the article is the founder of Terraform Industries, which aims to use solar power to synthesize natural gas. Likely the hope is to get efficiency by closely integrating the solar panels with the desalination plant, similar to how they do it in that project.

I think if they really can get cheap desalination, there are many cities that would buy the water and pay more than agriculture. It wouldn't make sense to tackle agricultural use first, since it's more price-sensitive.

> Given that California electricity is pretty expensive even though we have lots of solar energy, how can that be done? Why not sell it to the grid?

Look at wholesale prices of electricity on the grid, and it's really small.

High prices in California are due to PG&E not maintaining the grid and causing massive amounts of damages, which everybody else pays for. Or something. It's really hard to find out why our transmission and distribution prices are so high. And for that matter, generation costs on bills are super high despite people usually getting much cheaper generation prices if they have a Community Choice option that procures their own renewable generation.

In any case, if the desalination project was funded by their own off grid generation they'd have really cheap electricity.

Huh; so desalinated water costs between $0.001 to $0.008 depending on the cost of the power.

I pay about $0.0136 / gallon so almost 10x the cost of producing.

Seems like (for non-agricultural use) we will be able to afford desalination -- I guess the real issue will be agricultural uses which are predicated on free water.

Making desalinated water is relatively cheap. You can do it with a clear tarp and sunlight. Desalination at scale is less cheap. Almost all water is cheap.

One (there are several) problem with desalination is that it often has to compete with free.

In many parts of North America the water is actually free. Like New York City-- their water falls from the sky in upstate NY, for free, into reservoirs that are not free to maintain to be distributed by colossal public works projects which are 100% not free.

Pumping water, desalinated or not, to your house is not free.

Whoever is charging you $0.0136/gal for water may indeed (especially if you don't live in an arid region with strange water rights laws) be charging you $0.0136/gal to pump water to your house and $0.0000/gal for the actual water.

If they switch to a desalinated source it is highly likely they will charge you $0.0136/gal + ≥$0.0010/gal = ≥$0.0146/gal.

Of course, if you live near a source of salt water you can cut out the middleman and, after massive up-front capital outlays, desalinate your own water for ≥$0.001/gal assuming you live long enough to recoup the cost of the initial expense.

Does that price include wastewater treatment? In our old home waste water was billed separately but the usage was based on your water meter. So if you poured 1000 gallons into the storm drain you were charged 1000gals of fresh water + 1000gals of sewage. The freshwater cost almost nothing but treatment is where the real costs came in.
I don't think it's just the brine that's concerning. Remember all of those toxic pesticides and heavy metals the author mentions? The ones causing health problems for residents of the Salton Sea area?

Those have to go someplace as well, and they would be mixed into that brine. Dumping a more concentrated version of that sludge into the gulf sounds like a bad time for people who live in the gulf region.

Don't get me wrong, I love this idea. I just know that there are hidden externalities here which need to be examined and dealt with properly.

It's not a huge problem: it's the same salt that was in the sea, just more concentrated. Dilute it well enough and it's a literal drop in the ocean.

The article also mentions that inefficient thermal desal produces much more brine, with modern RO desal being more efficient.

Yes just dilute it. In freshwater. Freshwater you got from ... where already ?
Why do you want to dilute it in freshwater?
Could we just... eat it? A by-product sounds a lot more efficient to access than mining. And people pay a premium for "sea salt."
Yeah I don't think you can just dump it in the top end of the gulf of california, especially if that's where you're sourcing feedwater.
Why not?

Existing ocean currents in the Gulf of California driven by wind, tides, and thermohaline circulation are on the order of 10cm/s, tens of kilometers wide, hundreds of meters deep. Very little of that is associated with local rivers, which are relatively tiny because the whole region is extremely arid. Are you going to be desalinating 0.1 km^2 of water per second?

The Sun pumps a thousand watts a square meter onto the surface of the Earth on a 24-hour cycle and this tends to shake things up.

The brine doesn’t spread out automatically, it sinks, and it’s deadly to aquatic life.
It does spread out automatically. We call it diffusion.

Diffusion can also be assisted with very modest power requirements.

Pump it into the Salton Sea.
It already has too much toxic salt dust blowing around.
Love the vision. Feels like something that can’t get done and isn’t proposed as viable

- if a RE developer tried to do it for profit, they’d get shut down - ironically probably by environmental groups

- govt doesn’t actually build anything audacious anymore

Unless I’m missing something?

Actually, Arizona is quietly advancing a plan to do desalination with Mexico, and Arizona has a history of building audacious infrastructure (New highways, etc). We will probably see desalination but it won’t come out of California.
The CAPEX seemed a bit glossed over. $50ish billion isn't a small ask.

Perhaps I missed a zero somewhere, but 5MAF/yr would require pumping 6,900cfs of water to/from the gulf with 230+ft of head. We're going to need a bigger boat.

Maybe he’s hoping a well meaning billionaire will want to try it
I thought this was going to describe cloud seeding, artificially stimulating cloud formation: https://en.wikipedia.org/wiki/Cloud_seeding

It's been used in Dubai, the middle of the desert, to create fake rain (https://en.wikipedia.org/wiki/Cloud_seeding#cite_ref-8). It's been used in some US states (https://www.scientificamerican.com/article/eight-states-are-...).

Though it's effectiveness is debated and it cannot guarantee precipitation. It also has environmental impacts, possibly decreasing precipitation in other areas as well as causing health issues (it basically involves spreading salt or dry ice into the air).

It is a zero sum game (if it works). Taking moisture from the atmosphere in one place prevents it from precipitating someplace else.

I can envision a world in which wars are fought over this kind of behavior.

The reality is that 80-90% of all water used in California is used for agriculture, and about half of the produce is exported to other states and countries. All other uses of water including residential, commercial, golf courses, lawns and all the rest account for only 10-20% of the water used.

Sacramento, a virtual desert, is surrounded by rice fields, with huge exports to Japan.

Effectively California’s farmers are selling its water to other places in the form of produce.

There are 5 main water problems that affect California and the Southwest:

1. Increasing freshwater production along the coast.

2. Increasing freshwater distribution to Southwest regions.

3. Increasing stormwater drainage to prepare for more intense storms. This includes identifying and remediating areas of so-called "100 year" flood plains that put lives and property at risk.

4. Increasing freshwater storage capacity to prepare for longer droughts and lower rainfall.

I wonder how alternative energy sources would change the math on this project. Is there a small scale or phased approach possible where a desalination plant with 1/5th the final capacity is built first? Big massive, all at once projects seem more likely than not to fail. (High speed rail anyone?)
> Los Angeles, a thriving metropolis with a population and economy larger than Australia’s

When did LA become a city of 26 million?

California's population is larger than Australia's. But yeah, LA's specifically is about half (give or take)
This gives Southern California including LA a listing as a single supercity population 24.4 million which is close but not quite there.

https://en.wikipedia.org/wiki/List_of_megalopolises

That’s a goddamned stretch if ever there was one. SoCal is more or less a recognizable region in California, and does indeed have almost 24M people, but nobody sane would call it a megalopolis without some weird agenda I can’t quite figure out.

Los Angeles is a city of about 3.9M. LA County is well, apparently experiencing some population decline because last I checked this was a little over 11M, but apparently as of 2023 is around 9.6M. If you count all the urban areas that touch, are surrounded by or very very close by without driving out into the bloody desert, “greater” LA, 18.4M is probably a reasonably close figure. Some of what’s on Wikipedia’s map is a stretch though. The further East you go, the more rural California gets until you’re either in Clark County or Arizona.

San Diego though is by no means part of that. It’s a large city in its own right and going downtown to downtown, you’re talking at least between 2 and 2.5 hours driving. You can also just fly, on a regular commercial airline.

A megalopolis isn't characterized by a placename and an airport. It's characterized by continuous development. You can drive from Ventura to Yucaipa to to Temecula to Rosarito Beach in Mexico, and the only time you'd need to get a mile away from of high-density development is 5 miles of I-15 in a mountain pass, or 10 miles of coastal highway, both to span the Santa Ana Mountains (property of the US Forest Service / Cleveland National Forest for the most part).

The Los Angeles Conurbation is about to swallow the northern Imperial Valley as well, because the I-10 corridor is made up of flat sandy dirt that could be a suburban back yard.

> A megalopolis (/ˌmɛɡəˈlɒpəlɪs/) or a supercity,[1] also called a megaregion,[2] is a group of metropolitan areas which are perceived as a continuous urban area through common systems of transport, economy, resources, ecology, and so on.[2] They are integrated enough that coordinating policy is valuable, although the constituent metropolises keep their individual identities.[2]

2 hours driving, in the US, is a natural distance for complementary but distinct cities to form. NYC/Philly/DC, SF/San Jose, LA/SD.

Megapolis is the correct term.

How is Tijuana, Mexico part of the Southern California megalopolis? That doesn't make much sense and I wouldn't consider San Diego part of the LA/OC megalopolis either, especially since there's a mountain range separating them with not much civilization in between.

Either way, we've got enough people to invade Australia and no Kokoda Trail is going to help them this time! We must get a plan to Gavin Newsom’s desk at once. The Koala mines will be ours!

The only thing that stops San Diego from being connected to LA/OC is the Camp Pendleton Marine base for about 7 miles. Basically you can go from Tijuana all the way up the coast to LA and only have that stretch not populated with civilians.
San Diego and Tijuana used to be one region that was cut in half by the Mexican-American war of 1846.

https://en.wikipedia.org/wiki/Conquest_of_California

https://en.wikipedia.org/wiki/San_Diego–Tijuana

Australia pre infiltrated California with eucalyptus sleeper groves decades ago ... one tossed ciggie and Cali's toast.
I guess the huge wildfires in Australia a few years ago were a dry run?
Wikipedia for Greater Los Angeles gives 18.4 million, so closer to 70%.
Australia's population is now meant to be 27 million.

https://www.abc.net.au/news/2024-01-24/australias-population...

> Last year, the world installed about 437 GW of solar electricity, or roughly one acre of panels every 10 seconds

How can that be true.

That's over three million acres.

Really?

Sounds about right assuming the global solar installation figures are ballpark correct. (See World in data [1], etc).

We've gone from ~200 GW installed per year to ~400 GW per year.

The India mega solar farm [2] is quoted as designed to deliver 30 GW from 72,600 hectares (179,398.5 acres).

Sanity checking ( 437 / 30 ) x 179,398 gives 2.6 million acres.

[1] https://ourworldindata.org/grapher/installed-solar-pv-capaci...

[2] https://en.wikipedia.org/wiki/Gujarat_Hybrid_Renewable_Energ...

There was a four thousand acre solar panel setup in Texas that was demolished by a hail storm a month ago.

Despite the "everything in Texas is bigger" motto, worldwide 3 million doesn't sound too far fetched if a single install in Texas is already 4,000 acres.

That's surprising, I've seen lots of photos of solar farms post hail storm where there was minimal damage. Perhaps the hail is bigger in Texas (at least for that storm).
Reports put the size around that of a golf ball with some upwards of a baseball. Those larger ones definitely have the energy to wreck anything but the most resilient glass.

A similar event happened in Nebraska last year (I think). They got the plant back up and running in roughly six months, which is quite the feat given the amount of labor disposing and installing new panels involves.

Then again, six months is a very long time for a plant to be producing no power at all. Keeping these plants spread out will increase the odds of outages but will be necessary for them to become critical infrastructure.

The 437GW is accurate, but depending on how you measure the "acres" the MW/acre figures seems fairly low, they may be counting a lot of empty space on parcels that are not actually devoted to solar panels. Using estimates from Fig4 here, I'd expect roughly half as many acres as that to be occupied by actual solar panels:

https://www.energy.gov/sites/default/files/2022-01/lbnl_ieee...

Are you saying it isn't very much?

It's about equivalent to a single larger US county (but well down the list from largest).

Most people don't realize how much solar is getting deployed and how quickly it's grown, so the poster is most likely expecting the opposite direction.

Meanwhile, numbers in solar always seem to exceed estimates, see for example this veteran forecaster joking about the 1TW/year deployment mark:

> Reasons solar industry analysts don't want to forecast that a terawatt gets built in a year, ever:

> - axis on chart doesn't go that high

> - makes wind industry feel bad

> - can't find markets to put it all in without the number being clearly ridiculous for that specific market

https://x.com/solar_chase/status/1791088142738411565?s=46