Short Takes on Big Questions

Help restore my faith

Quick takes on a variety of topics I am skeptical about.

Can We Accomplish Much by Planting Trees?

To be clear: deforestation – cutting down existing forests – is a huge problem, for many reasons including but not limited to climate change, and we should definitely stop doing it.

It’s popular to talk about planting new trees as a way to reverse climate change. But it’s not clear to me that this can serve more than a minor role. There is a limited amount of suitable land; trees can absorb more heat than unforested terrain; they can burn, releasing their carbon back into the atmosphere. Here’s a skeptical article in Nature. Bill McKibben had a nice piece. To quote: “I fear I’m on occasion a literal tree-hugger … so it pains me a bit to say, massive tree-planting campaigns are under assault as a climate tool, and that the assault seems to have some real merit.”

Has anyone seen a sound analysis of the total amount of carbon sequestration (and net climate benefit) we can get from planting trees, using realistic assumptions? Note that, per my piece on Embracing Overshoot, a short-term win is enough to be interesting. If we can pull down some carbon for 30 or 40 years, that’ll give us time to develop longer-term solutions. But it’s not clear to me that planting new trees can move the needle even temporarily.

Where’s the Darned Storage?

It’s been clear for a long time that clean, reliable electricity requires long-term storage. In some regions, energy requirements are high during the winter, just when solar power is at a minimum. And both solar and wind can experience extended periods of poor availability, as Europe just learned. People expect high reliability from the grid, and we’re not going to get there without ample storage capacity. Lithium-ion batteries look good for overnight storage, but are completely impractical for filling in multi-day shortfalls, let alone seasonal imbalances.

Pumped hydro is nice, but there are only so many suitable sites. I’ve been reading about “flow batteries” for a long time, but where are the pilot projects? Using excess electricity to produce fuels such as hydrogen or ammonia also seems promising, but again, is anyone actually implementing this? Unlike (say) nuclear fusion, these don’t seem like exotic technologies, so I’m confused why there haven’t been visible pilot projects and I worry that it might be because the ideas are not actually practical.

Note that a long-duration storage system will only be used a limited number of times – as little as one cycle per year, unlike overnight battery storage which would be used daily. So it will need to be extremely cheap, in terms of dollars per kWh of storage capacity. Perhaps no one can make the economics work out.

(Update: there is at least one pilot project, of an iron-air battery, in the works. Hope it goes well! Also, by tying together grids from different regions, we might be able to reduce the need for storage; when it’s winter in one place, it’s summer somewhere else.)

Is Demand Management Really a Thing?

The previous section discussed long-term storage, but renewables also have problems with short-term reliability – many sources are intermittent. Solar power is only there when the sun shines, wind turbines only spin when the wind blows. They tend to complement one another, but sometimes they both fade at once. Demand management is the idea that when supply is low, we can reduce electricity usage to match, and thus minimize the need for batteries and other alternative sources.

When this idea comes up, the examples often involve household usage: “your dishwasher will wait to run when demand is low”, “your electric car will charge at night, when wind power peaks”. What happens if you forget to leave soap in the machine? What if you need to run an errand after dinner and your car hasn’t charged? People are quick to resent imposed inconvenience. Demand management is an elegant concept, but I don’t think it would take too many loads of unwashed dishes for the whole idea to become an SNL sketch and a bad memory.

I can imagine more promise in the commercial and industrial sector, where most energy use occurs, users are more sophisticated, and cost is more important. Thermal storage systems also seem interesting. But I haven’t seen a serious attempt to quantify the potential for demand management on a realistic basis.

Also note that supply shortages can be short-term (overnight), intermediate (a multi-day weather anomaly, see the first link in this section), or long-term (winter). Is there a plausible case for demand management that addresses intermediate or long-term shortages?

Also Nuclear, Will That Be a Thing?

I mean, I get it. The potential is enormous. It would be great to have more carbon-free baseline power. The safety concerns are massively overblown, coal kills way more people. The problem isn’t the technology, it’s over-regulation, and NIMBYs, and stupid over-sized over-complicated overly-custom mega installations. China and France plan to build more reactors. Plus nuclear still feels, you know, sciencey; I like that.

Bill Gates is doing nuclear. Y Combinator, too. Oh, and Peter Thiel, don’t want to be on his bad side.

But Small Modular Reactors have been a topic of conversation for an awfully long time now. NuScale started in 2007, and (per Wikipedia) “has been approved to build test reactors in Idaho, in 2029 and 2030”. TerraPower (backed by Bill Gates) started in 2008; their planned prototype reactor in China “[has] been abandoned due to technology transfer limitations placed by the Trump administration” (Wikipedia again). It’s just really hard to see us getting past “too many people hate nuclear and it’s too hard to get anything done” on a useful time frame.

As a bonus, apparently nuclear power requires more water for cooling than any other form of power generation, edging out coal. This might hamper our ability to dramatically scale up even if the other issues were addressed.

Finally, fusion. Even if I were to believe that “this time is finally different” and the current crop of fusion startups may finally achieve positive energy production… it’s not obvious that fusion solves any of the problems that have been holding back fission power. Fusion plants will still be large, complicated, take a long time to construct, need lots of water, etc.

Oil’s Long Tail

All of the plans and projections for phasing out fossil fuels revolve around green alternatives becoming cheaper, or at least cost-competitive. And there’s lots of very encouraging data on this topic.

However, once demand for fossil fuels enters a (hopefully!) steep decline, the law of supply and demand will push down prices. In the last 10 years, oil prices have mostly fluctuated in the $40 – $100 / barrel range (excluding a brief anomaly early in the Covid pandemic). But some oil fields would remain profitable at much lower prices. The first source I found regarding Saudi Arabian oil pegs the cost of production at $2.80 per barrel. If oil drops to, say, $10 / barrel, Saudi Arabia will have a hard time balancing their budget, but it will still be worth their while to keep pumping, and it will be hard to beat that remaining production on price.

I don’t know how big a problem this will be. But our plans and projections should take it into account.

Ending on a Positive Note

I’m skeptical about these specific topics, but I’m optimistic about our ability to address climate change. We just need to clear away the false solutions so we can focus on the ideas with more promise.

Also: I may easily be wrong about any or all of these. If I’m skeptical about X, and you’ve seen a concrete plan that lays out how X could work in practice, please let me know!