We need more water7th December 2024
Our civilization operates under the assumption that water is abundant. But rain and groundwater, the primary sources of freshwater, won’t provide enough for our growing needs. For the past two decades, we’ve been facing freshwater shortages worldwide.The changing patterns of rain make rain unreliable. Moreover, rain isn’t filling our groundwater reserves faster than we are extracting them. We will soon run out of groundwater just like fossil fuels[1]. We need to act now.Rationing water is a bad solution. Rationing forces people to choose the most economically productive use for the limited water available to protect their livelihoods. And growing food is not an economically productive use case for water. If we force this, we will cause food shortages, famines and hunger. Poorer people will be the first to suffer. So let’s not do it.Software that helps “manage” water won’t fix the problem if we ration. When we have a limited physical resource, software is in no way going to magically help us produce more from less. It’s like trying to boil an ocean with a candle. No matter how efficient we are, it’s simply impossible to “manage” or fix the problem. We need more water.
In the Industrial Sector, the water consumption per capita per year of the US is 765 m3, China 102 m3 and India 13 m3. In order to enjoy the standard of living like in the US, the rest of the world will need a lot more water (58x more in India’s case)[2].That’s a lot of gloomy talk. Fortunately, we have a way out.What’s the solution?97% of Earth’s water is seas and oceans. We are surrounded by so much of it; just not in the convenient freshwater form we want.Rain isn’t reliable. We’ve known that for centuries now. The way out of water shortage is by desalinating seawater using RO (reverse osmosis) filters—the ones used in water purifiers. This is the only reliable solution that works at scale.RO filtration works by pressurizing the seawater and passing it through an RO filter. Instead of building a giant SWRO plant, we can create units that are stacked in a plug-and-play fashion and leverage mass production in a factory to drive the costs down significantly and iterate on the design quickly while targeting the customers who can afford it including a good profit margin at each stage, aka, The Tesla Master Plan strategy.Design a small end-to-end unit. Deploy huge numbers of these units. Drive the costs down while doing so. Iterate to deploy much more. This avoids a huge CapEx (a SWRO plant can be operational even with one unit). This is the essence of Industrial Capitalism.How much does it cost?Historically, the bottleneck for seawater reverse-osmosis (SWRO) has been the cost of the energy needed to pressurize the water.
Solar is getting 30% to 40% cheaper each time the production doubles. This cheap energy source is what we’ll use to power desalination.Modern SWRO plants produce a cubic meter (m3) of water for 2.5 kWh of electricity. Let’s look at a model that produces 1b m3 of freshwater per year.1b m3 of freshwater is produced from 285 MW of electricity (1b m3 per year * 2.5 kWh of electricity/m3).We’ll assume the sun is up for 6 hours/day, so the utilization is 25%. 1140 MW (285 MW * 4 – 1 part for SWRO, 3 parts to charge batteries) of solar costs $1.14b. To run the plant for the whole day, we’ll use batteries with a capacity of 5.1 GWh (285 MW * 18 hours). At $181/kWh[3], the batteries cost $923m. A SWRO plant this size costs $3.5b to build.[4]So, solar (at 25% utilization) + batteries + SWRO plant of this size costs $5.57b in total for a throughput of 1b m3/year.If solar utilization increases, the quantity of batteries and solar PV needed decreases. We can save about $877m at 50% utilization (12 hours).Pricing desalinated water at $0.5 per m3, the revenue is $500m/year or slightly over 11 years of operation to cover the capital cost.Let’s compare the $0.5/m3 cost with the actual tap water costs across the world.[5]
The red dotted line represents $0.5/m3.All around the world, desalinated tap water beats existing tap water costs today. We can add very healthy margins on top too.ConclusionThe need for water is extreme and exists. The technology is here. The economics are very favorable. Execution is what’s left.
In the Industrial Sector, the water consumption per capita per year of the US is 765 m3, China 102 m3 and India 13 m3. In order to enjoy the standard of living like in the US, the rest of the world will need a lot more water (58x more in India’s case)[2].That’s a lot of gloomy talk. Fortunately, we have a way out.What’s the solution?97% of Earth’s water is seas and oceans. We are surrounded by so much of it; just not in the convenient freshwater form we want.Rain isn’t reliable. We’ve known that for centuries now. The way out of water shortage is by desalinating seawater using RO (reverse osmosis) filters—the ones used in water purifiers. This is the only reliable solution that works at scale.RO filtration works by pressurizing the seawater and passing it through an RO filter. Instead of building a giant SWRO plant, we can create units that are stacked in a plug-and-play fashion and leverage mass production in a factory to drive the costs down significantly and iterate on the design quickly while targeting the customers who can afford it including a good profit margin at each stage, aka, The Tesla Master Plan strategy.Design a small end-to-end unit. Deploy huge numbers of these units. Drive the costs down while doing so. Iterate to deploy much more. This avoids a huge CapEx (a SWRO plant can be operational even with one unit). This is the essence of Industrial Capitalism.How much does it cost?Historically, the bottleneck for seawater reverse-osmosis (SWRO) has been the cost of the energy needed to pressurize the water.Solar is getting 30% to 40% cheaper each time the production doubles. This cheap energy source is what we’ll use to power desalination.Modern SWRO plants produce a cubic meter (m3) of water for 2.5 kWh of electricity. Let’s look at a model that produces 1b m3 of freshwater per year.1b m3 of freshwater is produced from 285 MW of electricity (1b m3 per year * 2.5 kWh of electricity/m3).We’ll assume the sun is up for 6 hours/day, so the utilization is 25%. 1140 MW (285 MW * 4 – 1 part for SWRO, 3 parts to charge batteries) of solar costs $1.14b. To run the plant for the whole day, we’ll use batteries with a capacity of 5.1 GWh (285 MW * 18 hours). At $181/kWh[3], the batteries cost $923m. A SWRO plant this size costs $3.5b to build.[4]So, solar (at 25% utilization) + batteries + SWRO plant of this size costs $5.57b in total for a throughput of 1b m3/year.If solar utilization increases, the quantity of batteries and solar PV needed decreases. We can save about $877m at 50% utilization (12 hours).Pricing desalinated water at $0.5 per m3, the revenue is $500m/year or slightly over 11 years of operation to cover the capital cost.Let’s compare the $0.5/m3 cost with the actual tap water costs across the world.[5]The red dotted line represents $0.5/m3.All around the world, desalinated tap water beats existing tap water costs today. We can add very healthy margins on top too.ConclusionThe need for water is extreme and exists. The technology is here. The economics are very favorable. Execution is what’s left.- ✅ Market
- ✅ Technology
- ✅ Economics
- [1] This is happening today in most Indian cities. The freshwater needed for domestic purposes is transported by water tankers multiple times each day, not groundwater pumped up.
- [2] Water use in industry/Population in 2015 gives us per capita consumption.
- [3] The price of batteries has declined by 97% in the last three decades - Our World in Data
- [4] How much does seawater desalination cost per gallon? Table 1’s Sorek project
- [5] The Water Price Index (EUR)