You’re out of free articles.
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Sign In or Create an Account.
By continuing, you agree to the Terms of Service and acknowledge our Privacy Policy
Welcome to Heatmap
Thank you for registering with Heatmap. Climate change is one of the greatest challenges of our lives, a force reshaping our economy, our politics, and our culture. We hope to be your trusted, friendly, and insightful guide to that transformation. Please enjoy your free articles. You can check your profile here .
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Subscribe to get unlimited Access
Hey, you are out of free articles but you are only a few clicks away from full access. Subscribe below and take advantage of our introductory offer.
subscribe to get Unlimited access
Offer for a Heatmap News Unlimited Access subscription; please note that your subscription will renew automatically unless you cancel prior to renewal. Cancellation takes effect at the end of your current billing period. We will let you know in advance of any price changes. Taxes may apply. Offer terms are subject to change.
Create Your Account
Please Enter Your Password
Forgot your password?
Please enter the email address you use for your account so we can send you a link to reset your password:
When there’s no way out, should we go down?

On August 20, 1910, a “battering ram of forced air” swept across the plains of western Idaho and collided with several of the hundreds of small wildfires that had been left to simmer in the Bitterroot Mountains by the five-year-old U.S. Forest Service.
By the time the wind-fanned flames reached the trees above the mining town of Wallace, Idaho, later that day, the sky was so dark from smoke that it would go on to prevent ships 500 miles away from navigating by the stars. A forest ranger named Ed Pulaski was working on the ridge above Wallace with his crew cutting fire lines when the Big Blowup bore down on them and he realized they wouldn’t be able to outrun the flames.
And so, in what is now wildland firefighting legend, Pulaski drove his men underground.
Sheltering from a forest fire in an abandoned mineshaft was far from ideal: Pulaski held the panicked men at gunpoint to keep them from dashing back out into the fire, and he and the others eventually fell unconscious from smoke inhalation. But even now, more than a century later, there are few good options available for people who become trapped during wildfires, a problem that has caused some emergency managers, rural citizens, and entrepreneurs to consider similarly desperate — and subterranean — options.
“We have standards for tornado shelters,” Alexander Maranghides, a fire protection engineer at the National Institute of Standards and Technology (NIST) and the author of a major ongoing assessment of the deadly 2018 Camp fire in Paradise, California, told me. “We don’t have anything right now for fire shelters.”
That’s partially because, in the United States, evacuation has long been the preferred emergency response to wildfires that pose a threat to human life. But there are times when that method fails. Evacuation notification systems can be glitchy or the alerts sent too late. Roads get cut off and people get trapped trying to get out of their neighborhoods. Residents, for whatever reason, are unable to respond quickly to an evacuation notice, or they unwisely decide to “wait and see” if the fire gets bad, and by then it’s too late. “If you can get out, you always want to get out,” Maranghides said. “But if you cannot get out, you don’t want to burn in your car. You want to have another option and among them — I’m not going to call it a ‘Plan B,’ I’m going to call it the ‘Plan A-2.’ Because we need to plan for those zero notification events.”
One promising, albeit harrowing, option has been TRAs, or “temporary refuge shelters” — typically unplanned, open areas along evacuation routes like parking lots where trapped citizens can gather and be defended by firefighters. Hardening places like schools or hospitals so they can serve as refuges of last resort is also an option, though it’s difficult and complex and, if done improperly, can actually add fuel for the fire.
Beyond that, you start getting into more outside-the-box ideas.
Tom Cova is one such thinker. He has been studying wildfire evacuations for three decades, and when I spoke to him recently to discuss the problem of traffic jams during fire evacuations, he told me that in Australia, “they have fire bunkers — private bunkers that are kind of like Cold War bunkers in the backyard, designed to shelter [people] for a few minutes if the fire’s passing.”
Unbeknownst to me at the time, Cova has even gone on record to say he’d consider one for himself if he lived on a dead-end road in California’s chaparral country. “My family and I would not get in our car and try to navigate the smoke and flames with bumper-to-bumper taillights,” he told the Los Angeles Times back in 2008. “We would just calmly open up, just like they do in Tornado Alley — open the trap door and head downstairs. Wait 20 minutes, maybe less, and come back and extinguish the embers around the home.”
Get one great climate story in your inbox every day:
Bunker-curious Americans can get easily discouraged, though. For one thing, Australia’s bushfires burn through areas fast; if you’re sheltering underground from an American forest fire, you could be in your bunker for considerably longer. For another, there are very few guidelines available for such bunkers and as of yet, no U.S. regulations; even Australia, where there are standards, generally recommends against using fire bunkers except at the highest-risk sites. Then there is the fact that there is almost no existing American fire bunker market if you wanted to buy one, anyway. “You would think more people would have wildfire shelters, but they don’t,” Ron Hubbard, the CEO of Atlas Survival Shelters, told me. “Even when there was a big fire not that many years ago and all those people died in Paradise … it has never kicked in.”
Hubbard is technically in the nuclear fallout shelter business, though he’s found a niche market selling a paired-down version of his marquee survival cellar, the GarNado, to people in wildfire-prone areas. “How many do I sell a year? It’s not a lot,” he admitted. “I thought it would have been a lot more. You’d think I’d have sold hundreds of them, but I doubt — I’d be lucky if I sell 10 in a year.” Another retailer I spoke to, Natural Disaster Survival Products, offers “inground fire safety shelters” but told me that despite some active interest, “no one has bought one yet. They are expensive and not affordable for many.”
Installing a Wildfire Bunkerwww.youtube.com
Hubbard stands by his bunker’s design, which uses a two-door system similar to what is recommended by Diamond, California’s Oak Hill Fire Safe Council, one of the few U.S. fire councils that has issued fire bunker guidance. The idea is that the double doors (and the underground chamber, insulated by piled soil) will help to minimize exposure to the radiant heat from wildfires, which can reach up to 2,000 degrees. It’s typically this superheated air, not the flames themselves, that kills you during a wildfire; one breath can singe your lungs so badly that you suffocate. “Imagine moving closer and closer to a whistling kettle, through its steam, until finally your lips wrap themselves around the spout and you suck in with deep and frequent breaths,” Matthew Desmond describes vividly and gruesomely in his book On the Fireline.
This is also why proper installation and maintenance are essential when it comes to the effectiveness of a bunker: The area around the shelter needs to be kept totally clear, like a helicopter landing pad, Hubbard stressed. “You’d be stupid to put a fire shelter underneath a giant oak tree that’s gonna burn for six hours,” he pointed out.
If there is “a weakness, an Achilles heel of the shelter,” though, “it’s the amount of air that’s inside it,” Hubbard said. Since wildfire shelters have to be airtight to protect against smoke and toxic gases, it means you only have a limited time before you begin to risk suffocation inside. You can extend the clock, theoretically, by using oxygen tanks, although this is part of the reason Australia tends to recommend against fire bunkers in all but the most extreme cases: “Getting to a tiny bunker and relying on cans of air in very unpleasant conditions and being unable to see out and monitor things would be a very unpleasant few hours,” Alan March, an urban planning professor at the University of Melbourne, once told the Los Angeles Times.
Private fire bunkers, with their limited capacities, can start to feel like they epitomize the every-man-for-himself mentality that has gotten some wildfire-prone communities into this mess in the first place. Something I’ve heard over and over again from fire experts is that planning for wildfire can’t happen only at the individual level. NIST’s Maranghides explained, for example, that “if you move your shelter away from your house, but it’s next to the neighbor’s house, and your neighbor’s house catches on fire, preventing you from using your shelter, you’re going to have a problem.” A bigger-picture view is almost always necessary, whether it’s clearing roadside vegetation along exit routes or creating pre-planned and identifiable safety zones within a neighborhood.
To that end, bunkers are far from a community-level solution — it’s impractical to have a cavernous, airtight, underground chamber by the local school filled with 1,000 oxygen tanks — and they’re not a realistic option for most homeowners in rural communities, either. Beyond requiring a large eyesore of cleared space for installation on one’s property, they’re expensive; Hubbard’s fire shelter starts at $30,000, and that’s before the oxygen tanks and masks (and the training and maintenance involved in using such equipment) are added.
The biggest concern of all when it comes to wildfire bunkers, though, might be the false sense of security they give their owners. Evacuation notice compliance is already a problem for fire managers; by some estimates, as many as three-quarters of people in wildfire communities hesitate or outright ignore evacuation notices when they are issued, even when not immediately evacuating is one of the most dangerous things you can do. But by having a shelter in one’s backyard, people may start to feel overconfident about their safety and linger longer, or decide outright to “shelter in place,” putting themselves and first responders in unnecessary danger.
As far as Hubbard is aware, no one has actually ridden out a wildfire in one of his shelters yet (people tend to install them, and then he never hears from them again). But there have been reported cases of homemade fire bunkers failing, including a retired firefighter who perished in a cement bunker on his property with his wife in Colorado’s East Troublesome fire in 2020.
Even Pulaski’s celebrated escape down the mineshaft resulted in tragedy. Though the forest ranger is remembered as a hero for his quick thinking and the 40 men he saved from the Big Blowup, the stories tend to gloss over the five men who either suffocated or drowned in the shallow water in the mine while unconscious from the smoke.
Some things just don’t change over 100 years: You will always have the greatest chance of surviving a fire by not being in one at all.
Read more about wildfires:
Log in
To continue reading, log in to your account.
Create a Free Account
To unlock more free articles, please create a free account.
Citrine Informatics has been applying machine learning to materials discovery for years. Now more advanced models are giving the tech a big boost.
When ChatGPT launched three years ago, it became abundantly clear that the power of generative artificial intelligence had the capacity to extend far beyond clever chatbots. Companies raised huge amounts of funding based on the idea that this new, more powerful AI could solve fundamental problems in science and medicine — design new proteins, discover breakthrough drugs, or invent new battery chemistries.
Citrine Informatics, however, has largely kept its head down. The startup was founded long before the AI boom, back in 2013, with the intention of using simple old machine learning to speed up the development of more advanced, sustainable materials. These days Citrine is doing the same thing, but with neural networks and transformers, the architecture that undergirds the generative AI revolution.
“The technology transition we’re going through right now is pretty massive,” Greg Mulholland, Citrine’s founder and CEO, told me. “But the core underlying goal of the company is still the same: help scientists identify the experiments that will get them to their material outcome as fast as possible.”
Rather than developing its own novel materials, Citrine operates on a software-as-a-service model, selling its platform to companies including Rolls-Royce, EMD Electronics, and chemicals giant LyondellBassell. While a SaaS product may be less glamorous than independently discovering a breakthrough compound that enables something like a room-temperature superconductor or an ultra-high-density battery, Citrine’s approach has already surfaced commercially relevant materials across a variety of sectors, while the boldest promises of generative AI for science remain distant dreams.
“You can think of it as science versus engineering,” Mulholland told me. “A lot of science is being done. Citrine is definitely the best in kind of taking it to the engineering level and coming to a product outcome rather than a scientific discovery.” Citrine has helped to develop everything from bio-based lotion ingredients to replace petrochemical-derived ones, to plastic-free detergents, to more sustainable fire-resistant home insulation, to PFAS-free food packaging, to UV-resistant paints.
On Wednesday, the company unveiled two new platform capabilities that it says will take its approach to the next level. The first is essentially an advanced LLM-powered filing system that organizes and structures unwieldy materials and chemicals datasets from across a company. The second is an AI framework informed by an extensive repository of chemistry, physics, and materials knowledge. It can ingest a company’s existing data, and, even if the overall volume is small, use it to create a list of hundreds of potential new materials optimized for factors such as sustainability, durability, weight, manufacturability, or whatever other outcomes the company is targeting.
The platform is neither purely generative nor purely predictive. Instead, Mulholland explained, companies can choose to use Citrine’s tools “in a more generative mode” if they want to explore broadly and open up the field of possible materials discoveries, or in a more “optimized” mode that stays narrowly focused on the parameters they set. “What we find is you need a healthy blend of the two,” he told me.
The novel compounds the model spits out still need to be synthesized and tested by humans. “What I tell people is, any plane made of materials designed exclusively by Citrine and never tested is not a plane I’m getting on,” Mulholland told me. The goal isn’t to achieve perfection right out of the lab, but rather to optimize the experiments companies end up having to do. “We still need to prove materials in the real world, because the real world will complicate it.”
Indeed it will. For one thing, while AI is capable of churning out millions of hypothetical materials — as a tool developed by Google DeepMind did in 2023 — materials scientists have since shown that many are just variants of known compounds, while others are unstable, unable to be synthesized, or otherwise irrelevant under real world conditions.
Such failures likely stem, in part, from another common limitation of AI models trained solely on publicly available materials and chemicals data: Academic research tends to report only successful outcomes, omitting data on what didn’t work and which compounds weren’t viable. That can lead models to be overly optimistic about the magnitude and potential of possible materials solutions and generate unrealistic “discoveries” that may have already been tested and rejected.
Because Citrine’s platform is deployed within customer organizations, it can largely sidestep this problem by tuning its model on niche, proprietary datasets. These datasets are small when compared with the vast public repositories used to train Citrine’s base model, but the granular information they contain about prior experiments — both successes and failures — has proven critical to bringing new discoveries to market.
While the holy grail for materials science may be a model trained on all the world’s relevant data — public and private, positive and negative — at this point that’s just a fantasy, one of Citrine’s investors, Mark Cupta of Prelude Ventures, told me over email. “It’s hard to get buy-in from the entire material development world to make an open-source model that pulls in data from across the field.”
Citrine’s last raise, which Prelude co-led, came at the very beginning of 2023, as the AI wave was still gathering momentum. But Mulholland said there’s no rush to raise additional capital — in fact, he expects Citrine to turn a profit in the next year or so.
That milestone would strongly validate the company’s strategy, which banks on steady revenue from its subscription-based model to compensate for the fact that it doesn’t own the intellectual property for the materials it helps develop. While Mulholland told me that many players in this space are trying to “invent new materials and patent them and try to sell them like drugs,” Citrine is able to “invent things much more quickly, in a more realistic way than the pie in the sky, hoping for a Nobel Prize [approach].”
Citrine is also careful to assure that its model accounts for real world constraints such as regulations and production bottlenecks. Say a materials company is creating an aluminum alloy for an automaker, Mulholland explained — it might be critical to stay within certain elemental bounds. If the company were to add in novel elements, the automaker would likely want to put its new compound through a rigorous testing process, which would be annoying if it’s looking to get to market as quickly as possible. Better, perhaps, to tinker around the edges of what’s well understood.
In fact, Mulholland told me it’s often these marginal improvements that initially bring customers into the fold, convincing them that this whole AI-for-materials thing is more than just hype. “The first project is almost always like, make the adhesive a little bit stickier — because that’s a good way to prove to these skeptical scientists that AI is real and here to stay,” he said. “And then they use that as justification to invest further and further back in their product development pipeline, such that their whole product portfolio can be optimized by AI.”
Overall, the company says that its new framework can speed up materials development by 80%. So while Mulholland and Citrine overall may not be going for the Nobel in Chemistry, don’t doubt for a second that they’re trying to lead a fundamental shift in the way consumer products are designed.
“I’m as bullish as I can possibly be on AI in science,” Mulholland told me. “It is the most exciting time to be a scientist since Newton. But I think that the gap between scientific discovery and realized business is much larger than a lot of AI folks think.”
Plus more insights from Heatmap’s latest event Washington, D.C.
At Heatmap’s event, “Supercharging the Grid,” two members of the House of Representatives — a California Democrat and a Colorado Republican — talked about their shared political fight to loosen implementation of the National Environmental Policy Act to accelerate energy deployment.
Representatives Gabe Evans and Scott Peters spoke with Heatmap’s Robinson Meyer at the Washington, D.C., gathering about how permitting reform is faring in Congress.
“The game in the 1970s was to stop things, but if you’re a climate activist now, the game is to build things,” said Peters, who worked as an environmental lawyer for many years. “My proposal is, get out of the way of everything and we win. Renewables win. And NEPA is a big delay.”
NEPA requires that the federal government review the environmental implications of its actions before finalizing them, permitting decisions included. The 50-year-old environmental law has already undergone several rounds of reform, including efforts under both Presidents Biden and Trump to remove redundancies and reduce the size and scope of environmental analyses conducted under the law. But bottlenecks remain — completing the highest level of review under the law still takes four-and-a-half years, on average. Just before Thanksgiving, the House Committee on Natural Resources advanced the SPEED Act, which aims to ease that congestion by creating shortcuts for environmental reviews, limiting judicial review of the final assessments, and preventing current and future presidents from arbitrarily rescinding permits, subject to certain exceptions.
Evans framed the problem in terms of keeping up with countries like China on building energy infrastructure. “I’ve seen how other parts of the world produce energy, produce other things,” said Evans. “We build things cleaner and more responsibly here than really anywhere else on the planet.”
Both representatives agreed that the SPEED Act on its own wouldn’t solve all the United States’ energy issues. Peters hinted at other permitting legislation in the works.
“We want to take that SPEED Act on the NEPA reform and marry it with specific energy reforms, including transmission,” said Peters.
Next, Neil Chatterjee, a former Commissioner of the Federal Energy Regulatory Commission, explained to Rob another regulatory change that could affect the pace of energy infrastructure buildout: a directive from the Department of Energy to FERC to come up with better ways of connecting large new sources of electricity demand — i.e. data centers — to the grid.
“This issue is all about data centers and AI, but it goes beyond data centers and AI,” said Chatterjee. “It deals with all of the pressures that we are seeing in terms of demand from the grid from cloud computing and quantum computing, streaming services, crypto and Bitcoin mining, reshoring of manufacturing, vehicle electrification, building electrification, semiconductor manufacturing.”
Chatterjee argued that navigating load growth to support AI data centers should be a bipartisan issue. He expressed hope that AI could help bridge the partisan divide.
“We have become mired in this politics of, if you’re for fossil fuels, you are of the political right. If you’re for clean energy and climate solutions, you’re the political left,” he said. “I think AI is going to be the thing that busts us out of it.”
Updating and upgrading the grid to accommodate data centers has grown more urgent in the face of drastically rising electricity demand projections.
Marsden Hanna, Google’s head of energy and dust policy, told Heatmap’s Jillian Goodman that the company is eyeing transmission technology to connect its own data centers to the grid faster.
“We looked at advanced transition technologies, high performance conductors,” said Hanna. “We see that really as just an incredibly rapid, no-brainer opportunity.”
Advanced transmission technologies, otherwise known as ATTs, could help expand the existing grid’s capacity, freeing up space for some of the load growth that economy-wide electrification and data centers would require. Building new transmission lines, however, requires permits — the central issue that panelists kept returning to throughout the event.
Devin Hartman, director of energy and environmental policy at the R Street Institute, told Jillian that investors are nervous that already-approved permits could be revoked — something the solar industry has struggled with under the Trump administration.
“Half the battle now is not just getting the permits on time and getting projects to break ground,” said Hartman. “It’s also permitting permanence.”
This event was made possible by the American Council on Renewable Energy’s Macro Grid Initiative.
On gas turbine backorders, Europe’s not-so-green deal, and Iranian cloud seeding
Current conditions: Up to 10 inches of rain in the Cascades threatens mudslides, particularly in areas where wildfires denuded the landscape of the trees whose roots once held soil in place • South Africa has issued extreme fire warnings for Northern Cape, Western Cape, and Eastern Cape • Still roiling from last week’s failed attempt at a military coup, Benin’s capital of Cotonou is in the midst of a streak of days with temperatures over 90 degrees Fahrenheit and no end in sight.

Exxon Mobil Corp. plans to cut planned spending on low-carbon projects by a third, joining much of the rest of its industry in refocusing on fossil fuels. The nation’s largest oil producer said it would increase its earnings and cash flow by $5 billion by 2030. The company projected earnings to grow by 13% each year without any increase in capital spending. But the upstream division, which includes exploration and production, is expected to bring in $14 billion in earnings growth compared to 2024. The key projects The Wall Street Journal listed in the Permian Basin, Guyana and at liquified natural gas sites would total $4 billion in earnings growth alone over the next five years. The announcement came a day before the Department of the Interior auctioned off $279 million of leases across 80 million acres of federal waters in the Gulf of Mexico.
Speaking of oil and water, early Wednesday U.S. armed forces seized an oil tanker off the coast of Venezuela in what The New York Times called “a dramatic escalation in President Trump’s pressure campaign against Nicolás Maduro.” When asked what would become of the vessel's oil, Trump said at the White House, “Well, we keep it, I guess.”
The Federal Reserve slashed its key benchmark interest rate for the third time this year. The 0.25 percentage point cut was meant to calibrate the borrowing costs to stay within a range between 3.5% and 3.75%. The 9-3 vote by the central bank’s board of governors amounted to what Wall Street calls a hawkish cut, a move to prop up a cooling labor market while signaling strong concerns about future downward adjustments that’s considered so rare CNBC previously questioned whether it could be real. But it’s good news for clean energy. As Heatmap’s Matthew Zeitlin wrote after the September rate cut, lower borrowing costs “may provide some relief to renewables developers and investors, who are especially sensitive to financing costs.” But it likely isn’t enough to wipe out the effects of Trump’s tariffs and tax credit phaseouts.
GE Vernova plans to increase its capacity to manufacture gas turbines by 20 gigawatts once assembly line expansions are completed in the middle of next year. But in a presentation to investors this week, the company said it’s already sold out of new gas turbines all the way through 2028, and has less than 10 gigawatts of equipment left to sell for 2029. It’s no wonder supersonic jet startups, as I wrote about in yesterday’s newsletter, are now eyeing a near-term windfall by getting into the gas turbine business.
Sign up to receive Heatmap AM in your inbox every morning:
The European Union will free more than 80% of the companies from environmental reporting rules under a deal struck this week. The agreement between EU institutions marks what Politico Europe called a “major legislative victory” for European Commission President Ursula von der Leyen, who has sought to make the bloc more economically self-sufficient by cutting red tape for business in her second term in office. The rollback is also a win for Trump, whose administration heavily criticized the EU’s green rules. It’s also a victory for the U.S. president’s far-right allies in Europe. The deal fractured the coalition that got the German politician reelected to the EU’s top job, forcing her center-right faction to team up with the far right to win enough votes for secure victory.
Ravaged by drought, Iran is carrying out cloud-seeding operations in a bid to increase rainfall amid what the Financial Times clocked as “the worst water crisis in six decades.” On Tuesday, Abbas Aliabadi, the energy minister, said the country had begun a fresh round of injecting crystals into clouds using planes, drones, and ground-based launchers. The country has even started developing drones specifically tailored to cloud seeding.
The effort comes just weeks after the Islamic Republic announced that it “no longer has a choice” but to move its capital city as ongoing strain on water supplies and land causes Tehran to sink by nearly one foot per year. As I wrote in this newsletter, Iranian President Masoud Pezeshkian called the situation a “catastrophe” and “a dark future.”
The end of suburban kids whiffing diesel exhaust in the back of stuffy, rumbling old yellow school buses is nigh. The battery-powered bus startup Highland Electric Fleets just raised $150 million in an equity round from Aiga Capital Partners to deploy its fleets of buses and trucks across the U.S., Axios reported. In a press release, the company said its vehicles would hit the streets by next year.