Are we in for a termination shock?
6. Cloud math, missing sulfur, and adaptation solutions
Note: The views expressed here are the author’s own and do not reflect the views of Azolla Ventures or Prime Coalition.
Housekeeping
One of the joys of academia was getting to know a field in gory detail, and then making a small contribution to it. Every now and then, I’d sit back and marvel at the whole enterprise: the US government, acknowledging that free markets would fail to fund basic science, paid me and a few hundred thousand others to advance the knowledge of the species. After starting a new research project, I’d eventually get to the point where I was ready to run experiments that nobody else in the history of humanity had ever run, and it was up to me to report back what I found. The experience was at once both arcane and deeply meaningful.
One of the pains of academia was that it took forever to publish.
So, nearly five years after defending my thesis, I’m proud to finally share a duo of papers newly live in the journal Chemical Science:
Growth kinetics determine the polydispersity and size of PbS and PbSe nanocrystals
Persistent nucleation and size dependent attachment kinetics produce monodisperse PbS nanocrystals
If science translated into plain English is more your jam, I also wrote an explainer:
Now back to our regularly scheduled programming.
Making it rain
Here’s a model for how clouds work:
There is water vapor in the air.
When there’s enough water vapor in one area, it would prefer to condense into liquid water droplets (or ice, if it’s cold enough).
There are two ways to condense, water vapor into water droplets:
Nucleate new droplets from thin air. To accomplish this, there needs to so much excess water vapor in the air that it’s really unhappy as a gas and has no choice but to spontaneously form liquid droplets.
Condense droplets onto onto existing nuclei – dust and other cloud condensation nuclei. Water (and most other matter) strongly prefers this route.
If enough droplets form in one part of the sky, you’ve got yourself a cloud.
If the droplets grow large and heavy enough, they’ll start to fall out of the sky as rain, snow, sleet, or hail. Which of those you get depends on cloud temperature, ground temperature, cloud size, moisture content, and so on.
While we know how to control particle nucleation and growth in fields like semiconductors (see tweetstorm above), we’re still learning how to do this in clouds. Naturally, desert cities and farmers are leading the way. I think this is a field to watch – there’s a huge opportunity for innovation to improve lives and livelihoods as the climate changes. So the following article about a USDA research project caught my attention:
In mankind’s eternal quest to milk the clouds for rain, the latest and perhaps most promising technology involves a Cessna spray plane flying into clouds, releasing electrostatically charged water droplets.
The Weather Modification cloud seeding project is headed by engineer Dan Martin working out of the United States Department of Agriculture research unit at College Station, Texas.
In earlier work, Weather Mod induced rain by sending a plane into a water-laden cloud with one flare attached to trailing edge of each wing tip. The pair of flares were either silver iodine for the upper portion of the cloud where it’s colder, or calcium chloride for lower warmer parts of the cloud.
“That’s working really well. In west Texas we typically average 20 inches of rain per year. With Weather Mod, we’ve increased rainfall by 12 to 15 percent,” says Martin.
“That’s an extra two to three inches. It may not sound like much, but it makes a big difference to farmers and ranchers in the area. They’re growing forages, cotton, corn, soybeans and cattle of course. That additional rain is huge to them.” …
“We charge the droplet because we know the cloud has its own charge. We give our droplets an opposite charge so water in the cloud is attracted and bonds to it. Our charged droplet grows until it’s heavy enough to fall to Earth as rain.”
This is cool as hell. Charging up and spraying water into clouds turns out to A) reduce hail, B) increase rainfall, and C) even bring clouds back from the brink of evaporation. And it seems to work better and more cheaply than state of the art flares. From my vantage point, here’s what seems to be the fundamental difference between the two methods:
Flares create airborne dust, which increases the concentration of cloud condensation nuclei and lowers the barrier for water vapor to form droplets.
Charged water droplets similarly act as seed particles, but they also use their charge as an attractive force between existing droplets, which increases the growth rate of droplets.
Could this ever be an economic choice? Let’s sketch out a rough picture.
Here’s the plane the USDA used: the Air Tractor 402B, a crop duster that retails for $945,000. A quick google tells me that there are cheaper models out there, but this seems like a fine place to start. Let’s assume the plane has a lifetime of 30 years and will require 15% of its original cost per year in the form of operation and maintenance costs.1 With these assumptions, here’s the annualized cost of the plane:
Plane cost = ($945,000 / 30 yrs) + ($945,000 × 15%/yr) = $173,250/yr
The article mentioned that the spray system, inclusive of installation and pilot training, costs $40,000. Since this is specialized equipment, let’s assume a lifetime of 10 years and operations and maintenance costs of 20% per year. So effectively, the cost becomes:
Spray system cost = ($40,000 / 10 yrs) + ($40,000 × 20%/yr) = $12,000/yr
Plane + spray system cost = $173,250/yr + $12,000/yr = $185,250/yr
For the sake of simplicity, we’re going to assume that these are the only costs.
Now, how much extra rain are you getting in return? The article called out 12-15% increases to west Texas’ typical 20 inches of rain per year. I calculate:
20 in/yr * 15% = 3 in/yr
Now for the tricky part: how much area can one plane cover? Here’s a wild guess: assume coverage of 100 acres per flight and that each plane can cover five 100-acre regions. In other words, assume frequent re-seeding will be needed throughout the year, which restricts a plane’s coverage to just a couple fields. So, here’s how much rainwater you’re getting:
3 in/yr × 500 acres = 1,500 acre-in/yr = 125 acre-ft/yr
You might be wondering why I ended up at the strange unit of acre-feet. I agree it’s an odd one. But it’s standard in water circles, so we’ll speak the language. Let’s put cost and production together: how much did we pay for the extra water?
$185,250/yr / 125 acre-ft/yr = $1,482/acre-ft
Is $1,482 per acre-foot a good number? Let’s compare it to a few benchmarks:
Groundwater recharge in the western US: $90-1,100/acre-foot
Average price in Texas: $485.52/acre-foot
Reservoir expansion in the western US: $1,700-2,700/acre-foot
Seawater desalination: $1,900-3,000+/acre-foot
What I pay for water in Somerville, MA: $3,340/acre-foot (not counting sewer costs!)
So look, after reading an article and making some wild assumptions, it’s not totally clear how the economics will work. But this seems to be in the ballpark, and we have an abundance of everything needed – crop dusters, spray systems, jet fuel and so on. I think that in time, this might just pencil out for a lot of people. Keep in mind that climate change is making the atmosphere more humid. At the end of the day, it will come down to a fundamental question of climate adaptation: as the climate changes around us, what are we willing to pay for the resources we need?
Aerosols
When you burn fossil fuels, a few things happen:
The carbon becomes carbon dioxide,
The hydrogen becomes water,
The C-C and C-H bonds release lots of energy and powers our society,
And if conditions are right, whatever contaminants are in the fuel also get oxidized.
We’re going to take a look at one common contaminant: sulfur. When you burn it, you make sulfur oxides. These emissions are generally considered bad for the environment. They cause acid rain, irritate lungs and eyes, and damage ecosystems. So over decades, there’s been a big push to get sulfur out of fuels. For example, there’s a whole subfield of chemistry devoted to hydrodesulfurization – cleaving sulfur impurities out of organic compounds you might find in oil.2
The last bastion for high-sulfur fuel was the shipping industry, which powers its vessels with the heaviest, dirtiest fuel oils out there – about 3.5% sulfur by weight. But even this is changing, with the International Marine Organization banning high-sulfur fuel in 2020.
But sulfur also has an effect on the climate: cooling. Sulfur aerosols cause clouds to become more reflective, sending incremental sunlight back into space and preventing surface warming. Inspired by sulfur-rich volcanic eruptions that have temporarily cooled the planet, like that of Mount Pinatubo in 1991, sulfur aerosols are now one of the main proposals of the geoengineering community.
So, in an attempt to clean up fossil fuels, we’ve actually made them into better global warmers. We’ve traded an environmental problem for a climate problem. But it’s all a matter of degrees. How big a deal is this?
A scientist named James Hansen thinks it's a big deal:
James Hansen, a climate scientist who shook Washington when he told Congress 33 years ago that human emissions of greenhouse gases were cooking the planet, is now warning that he expects the rate of global warming to double in the next 20 years.
While still warning that it is carbon dioxide and methane that are driving global warming, Hansen said that, in this case, warming is being accelerated by the decline of other industrial pollutants that they’ve cleaned from it.
Plunging sulfate aerosol emissions from industrial sources, particularly shipping, could lead global temperatures to surge well beyond the levels prescribed by the Paris Climate Agreement as soon as 2040 “unless appropriate countermeasures are taken,” Hansen wrote, together with Makiko Sato, in a monthly temperature analysis published in August by the Climate Science, Awareness and Solutions center at Columbia University’s Earth Institute.
In other words, declining sulfate aerosols makes clouds less reflective, enabling more sunlight to heat the Earth. It’s worth mentioning that Hansen isn’t the only one worried about this.
This leads me to three places:
This seems bad. You hope they’re wrong, but the case looks compelling. We’re ramping down emissions of the very chemical that some hope to ramp up in order to cool the planet.
If true, it means that we’re running yet another unplanned climate experiment: what happens to the rate of global warming when you cut out sulfur emissions?
I’d submit that we’ve already been doing geoengineering for decades, intentionally and unintentionally. I count oil infrastructure, highway systems, forestry, agriculture among them. It turns out that removing sulfur from fuels might just be another in a long string of exhibits.
If there’s anything to this theory, I hope we’re able to learn how effective sulfur aerosols really are, and how we might go about injecting them into the atmosphere to counteract global warming while avoiding environmental and health problems. Are we in for a termination shock — a rapid increase in temperature because we’ve stopped geoengineering — sooner than we think?
The IPCC adaptation report – part II
Last time, we talked about an Intergovernmental Panel on Climate Change (IPCC) report called Climate Change 2022: Impacts, Adaptation, and Vulnerability.3 Instead of trying to condense everything in its 3,675 pages into a few sentences, we did the opposite. Took a fraction of a single claim in the report and spent 2,000 words trying to understand it. We ended up at a 14-term mathematical equation linking crop yield to climate change, and then marveled at how reality really is just a series of entries in spreadsheets when you think about it. It was fun!
But fundamentally, we’re solutions people around here. I’ve been arguing for a while now that, just like any form of change, climate change creates opportunity. So it caught my eye when the IPCC proposed and rated a list of climate adaptation solutions. Today, we’ll kick the tires. Let’s get into adaptation solutions.
Let’s look at that list, which for those following along at home is on page 22 of the Summary for Policymakers:
As usual, there’s a lot to unpack. Let’s start at the highest level. They see four big system transformations needed in order to adapt to climate change:
Land and ocean ecosystems
Urban and infrastructure systems
Energy systems
Cross-sectoral (my interpretation: “everything else”)
Within these, there are two dozen key risks/adaptation options and then eight categories on which these options were rated. So suffice it to say, we’re not gonna go through all of it. But a few things caught my attention:
There are only a few options marked both “high feasibility” and “high confidence” – just forest-based adaptation, resilient power systems, and energy reliability. In other words, I think the authors see lots of work still to be done.
The biggest gap in feasibility seems to be at the institutional level, as measured by eyeballing the average circle size.
By that same token, the best dimension for feasibility seems to be environmental.
The technological and economic dimensions, toward which I gravitate, are decidedly mixed.
One conclusion you could take away is that there are opportunities to improve the feasibility of solutions that are infeasible today. For example, “efficient livestock systems” looks pretty infeasible today. If you’re, say, an alternative protein entrepreneur, you might position your technology as a way to get around today’s inefficient livestock systems and therefore help humanity adapt to climate change.
A different conclusion you could take away is that there are opportunities to put our collective foot on the gas for solutions that are feasible today. For example, “resilient power systems” looks pretty strong. If you’re a grid operator, maybe this would convince you to double down on grid hardening measures like weatherization.
Yet another conclusion you could take away is that this chart isn’t very meaningful. It’s a statement of what is currently true, created by academics and aimed at policymakers. Maybe don’t read too much into it.
I tend to fall into the this third camp. After staring at this chart for a while and looking at some of the background material that went into it, I struggle to make this make sense beyond vague generalities. Even some of the generalities don’t really fit my understanding of the world. Let me be the first to say that I could be wrong about how the world works! But, for example, why does disaster risk management have a medium feasibility rating while resilient power systems has a high rating? Maybe we’re not perfect at disaster risk management, but many parts of the world have at least established systems – I’d wager roughly as many as have resilient electrical grids. There’s more we could get into, but I found this exhibit a bit disappointing.
However, there’s a silver lining: later in this section, there are specific acknowledgments of tradeoffs between mitigation and adaptation. I think this is valuable because climate adaptation is just our latest attempt to engineer our surroundings to suit our needs. Not everything is going to be a win-win. When you build a seawall, you’re going to disrupt a coastal ecosystem and liberate some carbon. We’ve got to grapple with that sooner rather than later. When the time comes, will we be ready? I hope so.
Elsewhere:
Scaling Carbon Removal by Neil Hacker – really excellent work illustrating what it will mean to build a carbon removal industry.
Thanks for reading!
Please share your thoughts and let me know where I mess up! You can reply directly, leave a comment, or find me below:
LinkedIn (if you’d like to connect, let me know this is how you found me)
This is big business to oil companies. The scale of sulfur removal from oil is so large that it’s actually where we get most of our sulfur, which ends up in everything from rubber to semiconductors.
There’s a newer IPCC report that came out this month called Climate Change 2022: Mitigation of Climate Change. We’re not talking about that one.