Nov 18 2022

The Potential of Geothermal Energy

As we discuss the optimal path forward for the next 30 years to get to net-zero carbon emissions for the energy sector, one big variable is the real-world potential of geothermal energy. Right now in the US geothermal produces 0.4% of our electricity. That is almost negligible, and is not going to help get us to our goal without an order of magnitude or more increase. What is the probability that we can bring significant geothermal online within 20-30 years?

Producing electricity at large scale is mostly about turning turbines, which rotates a magnet within a coil of conducting cable which generates electrical current in the wires. Turbines are turned by two basic methods – mechanical or with steam which in turn is generated by some heat source. Hydroelectric and wind turbines rotate the turbines through mechanical power. Burning fossil fuel or nuclear power plants produce heat to create steam. Solar photovoltaics are the exception because they directly turn sunlight into electricity through the photoelectric effect. But direct solar capture can use sunlight to once again heat a target, create steam, and turn a turbine.

Geothermal energy uses steam created by the natural heat below the surface of the earth to turn a turbine to make electricity. In a recent TEDx talk, Matt Houde who is the cofounder of a geothermal energy company points out that there is enough heat in the ground to power our world for a billion years. It’s a practically unlimited energy source. Why isn’t that, then, problem solved – all the energy we can need for the foreseeable future (arguably longer than human civilization is likely to last on earth) is right beneath our feet? The problem is – that heat is hard to get to.

From my reading it seems that there are three types of geothermal energy depending on our ability to access the heat. Current geothermal, the kind making up that 0.4%, takes advantage of natural hot spring that reach near or at the surface. Boise Idaho, for example, directly heats building from natural hot springs. You can also use near surface heated water to create electrical power. This was the low-hanging fruit of geothermal, but if we want an order of magnitude increase we need to develop what is called advanced geothermal. This approach uses technology developed by the fracking industry to drill down to the heat, inject water if necessary (if water is not already present), and then use that heated water to drive turbines.

What is the potential of advanced geothermal? I read different estimates, but they range from 4-6% of current US electricity production by 2050. Here is a map (also pictured above) of the US showing the temperature at 7 km. Current drilling technology is good to get down to about this depth but not beyond. But in the 3-7 km range, drilling is expensive and takes time. Even if we take the high end of the estimates, 6% of energy production in the US by 2050 is nice, but also is not going to be a major solution for decarbonization. Can we get beyond that.

That is what Houde hopes to do. His company is developing a microwave drilling technique that vaporizes rock as it goes. The problem is, this technology is still in the lab. He hopes to drill his first holes in the next few years, advancing the technology in order to allow for deeper and deeper drilling. He wants to get to 10 km and beyond. Every kilometer deeper extends the map of potential geothermal energy locations. The deepest hole drilled so far is the Kola Borehole, which reached 12 km deep and took almost 20 years. Obviously we need vast improvements on this technology. While this project has potential, this is partly why thinking about the future of geothermal can be such a wildcard. How will this technology do, how long will it take, and how much will it cost?

I think the bottom line is that we cannot rely on the deep drilling technology that is only now currently in development, certainly not by 2050. Even advanced geothermal is still largely a future technology, but is farther along. If the estimates are accurate, we can probably count on geothermal for about 4% or so of our electricity needs by 2050, a 10-fold increase over todays 0.4%. If we include this in an assessment of the potential of each power source, what do we get?

Hydroelectric power is currently at about 6% in the US, with a potential to increase to 9% by 2050. That assumes static demand, which is not likely to be true, so these figures will likely be lower. In fact, hydro capacity may only increase enough to keep it at 6% or so as demand increases. Wind and solar have the greatest potential for significant growth. Wind is currently at 9.2% and solar at 2.8%. There are two issues with these sources, they are land-use intensive, and they are intermittent. Without getting into the same debate about the upper limit of wind and solar, let’s just say there is a range of estimates where they can get by 2050.

If we put solar panels on every rooftop in the country that would cover 30% of electricity demand. That would avoid the land-use issue. But 100% rooftop potential is unrealistic – let’s say we get to 66% of the potential, so that solar generates 20% of our electricity. If we also strongly encourage rooftop solar installations to have battery backup, that would go a long way to mitigate some of the intermittency. It could at least shift peak production to peak demand in the same day, which is huge.

Wind is less problematic as an intermittent source because if you spread it out over a reasonably large area, the intermittency tends to average out (unlike sunshine). But we still have the land use issues. Perhaps the best potential for wind is offshore, which is more efficient anyway and also mitigates a lot of the land use problem. The potential of wind depends upon how far into less and less optimal locations we are willing to push it. But projecting 30% wind by 2050 is within reason. This will require grid updates and some grid storage. I know some will deem these projections way too optimistic, others too pessimistic, but that’s the point – to take a middle approach. In any case, this is a thought experiment using average projections.

If, then, by 2050 we have 6% hydro, 4% geothermal, 20% solar, and 30% wind, that leaves 40% unaccounted for. Right now nuclear is 18.9% and fossil fuel is 61%. Many of our nuclear power plants are set to retire so that 18.9% can decrease if we don’t extend their lifespan and replace them. This is why I say that our real choice for the next 30 years is between nuclear and fossil fuel. Whatever percentage of that remaining 40% is not nuclear is going to be mostly fossil fuel (there is a tiny percentage of other energy sources, like biomass, which is unlikely to change). You can play with the numbers if you like, but this is as realistic a projection as I can put together from combining many sources. If you think wind and solar percentages will be lower, then we need even more nuclear. Hydro and geothermal are not likely to be dramatically different than those projections.

At the other end, if you think wind and solar can be 80-90%, that is so-far a completely unproven projection. Every single time people try to counter that statement with an example, they give an example of a subset of a grid (like the UK or Germany or California) and miss the point that these locations make high penetration work by using other points on the same grid to balance their supply and demand. That doesn’t count, because we need to make it work for an entire grid. Also, high penetration projections often assume massive grid storage. While we are working on this, we cannot assume this will work out. A big point is, we won’t have time to course correct if in 20-30 years we realize that grid storage is going to take a lot longer than we thought. What this means is that we have to start building nuclear now, even “warp speeding” nuclear by streamlining regulations. If we can get significant new nuclear online in 10-30 years, they will replace fossil fuel plants and hold us over until deep drilling for geothermal comes online, or massive grid storage, or even fusion. These are more realistic for the second half of this century, not by 2050.

The IPCC agrees – there is no solution without nuclear. The numbers simply don’t work. Any other projection is incredible wishful thinking that we simply cannot risk. Otherwise that 40% or so will be provided by fossil fuel, and we will blow past 1.5 C.


No responses yet