What would the pitfalls be of pumping sea water inland to refill aquifer

September 27, 2015 9:49 AM

No, not *directly* into aquifers, but onto the ground at a slow rate to let water slowly seep down, using the soil and rock strata to remove impurities, such as (hopefully) saline content.

IANA environmental engineer, but my inspiration comes from two places: in fifth grade, we had a lesson about using natural materials (rock, moss, dirt) to filter water. I wish I could remember more, but that lesson totally blew my mind and earned me one of my few Bs in that class. The other is the Saltwater spills associated with oil drilling in North Dakota. Farmland is being ruined, which makes me suspect salt is being deposited in the top soil. Apparently they are just dumping this water into salt water wells.

Anyway, so you can see where this is going: if we can pump oil across the US, surely we can make a slow flow pipeline a few hundred miles inland to CA's central valley. Buy one of those plots that have the signs "this field is bare because congress sucks" (paraphrased), hollow out a pit, lay some fill rock or crushed concrete, and create a slowly draining, slowly filling salt marsh, that hopefully results in desalinated water dripping into an aquifer.

So what are the other glaring problems with doing this?

posted by rubah to Science & Nature (21 answers total)

Soil salinity
posted by blob at 10:01 AM on September 27, 2015 [5 favorites]

You seem to misunderstand what will happen to the salt. It isn't going to get filtered out.
posted by Good Brain at 10:12 AM on September 27, 2015 [10 favorites]

Because that's not how this kind of thing works. Marshland can filter water of some contaminants, especially things like fertilizers and small amounts of oil, like what washes off of roads(small amounts of oil can be metabolized by soil bacteria, excess nutrients are soaked up by marsh plants). Salt would just get pulled down into the aquifers along with the water. In fact, the soils in the Central Valley are already suffering from salt in the aquifers, because the water pumped from deep wells is already slightly saline, and when it is sprayed on the fields, the water evaporates away, leaving the salt, which kills plants.
posted by rockindata at 10:16 AM on September 27, 2015 [2 favorites]

I am not a geologist, but I took a hydrogeology class in college. We spent the entire term studying aquifers, and one of the first things we learned was: once an aquifer is used up, you can't just re-inflate it by adding water, like you could with a dried out sponge. Most aquifers are layers of rock that are either very porous, like sandstone, or they are full of tiny cracks. Once you start pulling water out of those aquifers, the pressure from the rock layers above it makes the aquifer compress and condense. That process continues as long as you keep pulling water out of it. By the end of the process, all of the places where water once hid have been compressed, and if you try and force water back into it, it usually has nowhere to go.

Also, most aquifers formed over a geological time span, so you would need thousands of years to refill the ones that might be able to be refill, not decades or even centuries.
posted by colfax at 10:27 AM on September 27, 2015 [9 favorites]

I don't believe the biggest issue is desalination - for the most part this is a solved problem. I think the larger issue is transport costs. I would be shocked to find out that that the cost of transporting a liter of water 500 km (w/o gravitational assist) is lower than the cost of desalinating 1 liter.

We use oil pipelines because the cost of transport is much less than the cost of the oil itself. The artificial water economy we are trying to maintain (i.e. water is free) means transport is much more expensive than the water being transported. Some time in the future we will start to charge for the actual cost of water and these problems will get sorted out - but not without some hardship in between...
posted by NoDef at 10:31 AM on September 27, 2015

The Salton Sea is an accidentally created example of what you're describing, albeit with fresh water, but the example is useful. The Colorado River was accidentally diverted to a flat-lying area, creating a giant lake.

If you pour water onto land, you get a lake, and that lake, with no outflow, turns to salt water over time.
posted by Cool Papa Bell at 10:33 AM on September 27, 2015 [2 favorites]

Just to follow up:

Cost to transport oil by pipeline (probably similar for water): $5/barrel = ~ 9 cents/gallon
Cost to desalinate water: $3/ 1000 Gallons = ~ 0.3 cents/gallon

So transport costs are 30x the desalination costs.
posted by NoDef at 10:39 AM on September 27, 2015 [5 favorites]

Thank you, NoDef. This is a question that begged actual data.
posted by IAmBroom at 11:00 AM on September 27, 2015

Pumping costs are not the issue. Comparing to oil pipelines is misleading. Oil is very high viscosity and must always travel within pipes. That is not true for water.

We know precisely how much it costs to transfer water for hundreds of miles because Los Angeles already does it. It transfers water from near Sacramento at sea level to Los Angles again at sea level. This includes a pump lift of 2000 feet across the Tehachapi Mountains. Theoretically, if you wanted to build the aqueducts for sea water, you could do the same in reverse. Los Angeles gets water for less than a penny a gallon. Reversing the flow would be in the same ballpark since you are going from sea level to sea level.

The real reason this wouldn't be practical was discussed above. Seepage doesn't eliminate salt contamination. It simply poisons the soil. Cost is not a factor.

If you were to pay the cost of desalination, then you wouldn't pump the costly water back into the ground. You would simply use the desalinated water directly and thereby reduce dependence on aquifers.
posted by JackFlash at 11:48 AM on September 27, 2015 [6 favorites]

Cost to transport oil by pipeline (probably similar for water): $5/barrel = ~ 9 cents/gallon

JackFlash just beat me to it, but yes, this is the wrong comparison for many reasons, among them that a leaking water pipe is not the environmental catastrophe that a leaking oil pipeline is, making operations much cheaper and easier.

There are a lot of aquifer recharge projects going on, though obviously using fresh water for the reasons outlined above. Some use infiltration basins similar to what you describe, and others pump the water down wells directly into the aquifer.
posted by Dip Flash at 11:56 AM on September 27, 2015

We can desalinate seawater, but not in the manner that you describe.

The laws of Thermodynamics are involved here. The energy value of pure water and crystalline salt is higher than that of an equivalent amount of salty water. Which is to say, when the salt dissolves, it releases energy. To separate it out again takes energy; no passive system can do it. (That would be "perpetual motion.")

It can happen "in the wild", with the energy coming from sunlight. A large lake full of salty water can dry up, yielding rock salt. That's where the Bonneville Salt Flats came from, for instance. A lot of that has happened around the world over the course of geological history. (One of the coolest examples is the layer of rock salt under the Mediterranean which was created during a period when Gibralter was closed and the Mediterranean mostly dried up.)

But the process is very slow (in some cases taking tens of thousands of years) because it involves evaporating all the water off, which makes it useless for your purposes because it's the water you want.
posted by Chocolate Pickle at 12:56 PM on September 27, 2015 [1 favorite]

I agree comparing to an oil pipeline is rough, but when you can't just use gravity (i.e. a river) is it really off by more than a factor of 100? If I take the wikipedia numbers for the California project highlighted above:

3 x 10^9 m^3 of water transported per year
5000 GWh of energy consumed (11500 consumed - 6500 produced)

I still get something like .1 cent per gallon in energy costs alone - this does not count operational costs or infrastructure. Again compared to .3 cent per gallon for desalination the transport is not negligible. [These are all rough numbers, I didn't spend any time beyond a quick search - but even municipal water rates seem to be in this ballpark]

If it were cheap and easy to transport water, then we would have long distance (1000 km) water pipelines on a grand scale all over the world - in fact we do not. Instead where water is scarce we have local desalination plants or local treatment plants serving local needs. Both desalination and treatment are much more efficient and cost effective than transport.

My only point above was that even if desalination worked as described in the question through soil percolation, etc. and was thus essentially free, the idea still does not make financial sense due to the cost of transport.
posted by NoDef at 2:26 PM on September 27, 2015

NoDef, using your numbers, the California water project delivers water at an energy cost of 1.7 KWh per cubic meter. Desalination energy costs are triple that amount even with today's best desalination technology, let alone 50 years ago when the project started.

The California water project annual cost, including energy, bonds for infrastructure, labor and maintenance, delivers water for about 20 cents a cubic meter, less than half the cost of the best desalination. It seems that long distance transport is still significantly cheaper than local desalination, but perhaps getting closer.

Incidentally, that is less than 0.1 cent a gallon, not even in the same ballpark as pumping oil.
posted by JackFlash at 3:37 PM on September 27, 2015 [1 favorite]

The energy value of pure water and crystalline salt is higher than that of an equivalent amount of salty water. Which is to say, when the salt dissolves, it releases energy. To separate it out again takes energy; no passive system can do it. (That would be "perpetual motion.")

So, here's a proposed "perpetual motion" desalination proposal. Tell me why this wouldn't work on theoretical grounds:

Salt water is denser than fresh: sea water is around 1.027g/cm³ while fresh water is closer to 1.000g/cm³. So a column of sea water of any given height should exhibit 2.7% higher water pressure at the bottom than a column of fresh water of the same height.

The pressure at the bottom of a column of water of height h metres is approximately 0.1h bar.

Reverse osmosis desalination requires a pressure differential across the filtration membrane of around 40 bar. That's 2.7% of about 1500 bar, which is the pressure you'd find at the bottom of a 15km water column.

So attach an RO membrane to the bottom of a 15km pipe, and fill that pipe with fresh water as you sink it into the sea. Once it's sunk to maximum depth, the pressure difference between the water column inside the pipe and the sea outside should be enough to overcome the osmotic pressure across the membrane and produce a net flow of fresh water into the pipe; you should then see fresh water flow out of the top.

Leaving aside for a moment the fact that the ocean is not in fact 15km deep anywhere on Earth: where, in this scheme, is the desalination energy coming from?
posted by flabdablet at 5:27 AM on September 28, 2015 [1 favorite]

The Sahara Forest Project is looking at using seawater and concentrated solar power to reclaim desert land. But they handle the desal through solar power, which happens to be abundant in the vicinity.
posted by sagwalla at 5:48 AM on September 28, 2015 [1 favorite]

There is promising research being done in California on how to replenish aquifers, though as already noted above, the ground often subsides as you deplete an aquifer, so there is less and less holding capacity to replenish. It sounds like it might be more a technique to prevent aquifers getting over-depleted in the first place.
posted by anonymisc at 11:18 AM on September 28, 2015

Leaving aside for a moment the fact that the ocean is not in fact 15km deep anywhere on Earth: where, in this scheme, is the desalination energy coming from?

Ooh, neat puzzle. If you let the system run its course, I'm thinking it would move towards an equilibrium where the RO stops working because the exterior water column becomes graduated with high-density water and mineral precipitate at the bottom (increasing the pressure differential needed for further RO) and low density water at the top (reducing the pressure differential available) .
If so, then it's not a perpetual motion machine; the proximal energy is gravitational potential energy, and it will only work for as long as the heavier parts have somewhere lower to fall down to.
But... if you keep stirring up the oceans and constantly remixing the water (eg the existing actions of solar and tidal/orbital energies, some geothermal too), then perhaps it might never be able to reach that equilibrium, and the RO could keep going as long as the sun shines, the moon orbits, and Earth's nuclear sump sizzles.
posted by anonymisc at 11:58 AM on September 28, 2015

Best answer: in fifth grade, we had a lesson about using natural materials (rock, moss, dirt) to filter water

Somewhat repeating what others have said, the mistake you're making is that the above process [filtering] can purify water of impurities that are in suspension, but dissolved salts are not in suspension.

It is possible to remove dissolved salt via mechanical means [i.e. reverse osmosis], but that requires specialised membranes and energy.

The only other options for dealing with dissolved impurities are chemical/biological processes.
posted by HiroProtagonist at 9:38 PM on September 28, 2015

the RO could keep going as long as the sun shines, the moon orbits, and Earth's nuclear sump sizzles

which is motion as near to perpetual as makes no difference.

Interestingly, the underlying osmotic pressure for sea vs fresh water is only about 27 bar; with better membranes, this harebrained scheme might actually just about work for a while, given a dip pipe into the Challenger Deep.
posted by flabdablet at 10:58 PM on September 28, 2015

Response by poster: I was always hazy about the defining line between solution and suspension. Thanks HitoProtagonist! That actually makes the old assignment make more sense.

Also appreciate the link to Sahara Forest Project, sagwalla
posted by rubah at 10:12 PM on October 14, 2015

I was always hazy about the defining line between solution and suspension.

That's healthy, because the line itself is hazy. The substances that straddle it are colloids.
posted by flabdablet at 8:11 PM on October 15, 2015