BARCELONA, Spain — The microbes have assembled themselves into the rough shape of an open hand, reaching out from fluid at the bottom of the bottle. A dozen such containers sit side by side in the beachfront Barcelona Biomedical Research Park, each containing translucent, moss-colored filaments that cling to the glass. But I’m stuck staring at this one sample that has built something so solid-looking out of these weightless strands — its thick wrist supporting an upturned palm that dissolves, where the fingers should be, into curls of soft hair.

This is not normal behavior for these bacteria, explains Núria Conde Pueyo, a post-doctoral researcher at the European Molecular Biology Laboratory. Microcoleus vaginatus is a photosynthetic bacterium that lives in dry soils and, when grown in test tubes, tends to just float around in its nutrient-rich liquid. No one in this lab has seen it create this kind of elaborate architecture. Conde Pueyo shrugs. “We are trying to figure out why.”

The answer most likely lies in the recent regime of torture this lab has overseen. Three years ago, these bacteria were living in Alicante, an area on Spain’s Mediterranean coast, in a patch of arid dirt that’s turning into a desert. From these unpromising beginnings, Conde Pueyo subjected the soil samples to a process known as hardening, a sort of trial by fire to replicate the tough conditions these microbes are expected to face as our planet’s climate changes. They were made “very dry, very hot,” she explains, and subjected to “a lot of ultraviolet, some cycles that were very harsh.” She has taken to the job of tormentor with apparent glee: “I mistreat that soil a lot,” she smiles.

So far, Conde Pueyo has whittled down an initial 70 species of photosynthetic bacteria from the sample to just 16, including the M. vaginatus now cultured in the flasks. “They are the survivors,” she says. But the process doesn’t just sort the microbial men from the boys; the bacteria become tougher through their ordeal. The UV blasts and drought-like conditions force the microbes to make physiological and biochemical adaptations and perhaps even trigger mutations. Maybe that’s what’s made this one strain into builders. “I think it’s because we’ve changed the genome,” she suggests. 

The point of this experiment is to alter the course of life on our planet, explains Ricard Solé, the polymath physicist who heads up this research effort. Hardening is “the first phase of our terraformation experiments,” says Solé, who founded the Complex Systems Lab at Barcelona’s Universitat Pompeu Fabra around 35 years ago. Solé is frank about the towering ambition of the work, which aims to identify the key players in the soil, decode their genetics and edit their DNA, to provide an organic alternative to geoengineering. His lab has already shown it is possible to give these microbes a kind of ecological superpower: a tweak via CRISPR gene-editing can enable them to capture water in soils by transforming surrounding moisture into a trail of hyaluronic acid, a type of naturally occurring chemical goo used by the cosmetics industry. After hacking Conde Pueyo’s survivors, these tiny ecological engineers should have the power to work their way through the soil, capturing mist and hydrating the environment as they go. For regions like Southern Spain, on the brink of turning dusty and desiccated, this could be enough to redirect their trajectory, from ecosystem collapse to stability and rebirth.

The team is aware of how this plan might sound. People’s usual reaction is “this is dangerous,” says Solé, sitting in his lab decorated with posters of Frankenstein author Mary Shelley and Blade Runner’s Rick Deckard. Since the 1980s, advances in genetic engineering have generated feverish hopes as well as fierce backlash, resulting in an effective veto on introducing genetically modified organisms into wild ecosystems. 

By now, we all know what he calls “the unintended consequences story”: a newly introduced microbe could unleash cascading impacts, unforeseen knock-ons and extinctions rippling through the living fabric of Earth’s biosphere. “People say: ‘This will be catastrophic. This will destroy the planet’ with no evidence,” he says. A few years ago, Solé’s team decided to put the theory to the test by constructing a hyper-realistic artificial ecosystem and introducing various microbes — hardened or edited — to see how genuine, real-world mutants impact the whole system.

Since the 1980s, advances in genetic engineering have generated feverish hopes as well as fierce backlash, resulting in an effective veto on introducing genetically modified organisms into wild ecosystems.

There is a chance that these newcomers will multiply rapidly out of control, outcompeting all the existing bacteria and causing this simulated dryland ecosystem to collapse. Like Shelley’s monster, these mutant microbes could run amok, leaving these hubristic scientists to lick their wounds and clear up their mess. “Our prediction is that’s not going to happen.” But now it’s time to find out. 

Microscopic Rainforests

Drylands — arid environments ranging from inhospitable deserts to sunbaked savannas  — make up more than 40% of terrestrial Earth, an area approximately six times the size of Canada. Yet, compared to rainforests and reefs, these parched places are often viewed as degraded or relatively simple ecosystems, historically overlooked by many naturalists who assume scant water and volatile temperatures mean a meager existence for any inhabitants. But look closely at the terrain, and you’ll find splashes of orange, purple, white and green that bubble into lumps or extrude into centimeter-scale pinnacles. This remarkable “living skin” is made of microbes that weave together particles of sand and dirt into structures known as  biocrusts or cryptobiotic soil. Under the microscope, we can see it as a writhing network of densely packed fungi, mosses, lichen and bacteria. 

It takes a while, walking around the experimental site on the outskirts of Alicante, to see signs of this microbial life. To make sense of this flat blanket of tie-die scrub, Conde Pueyoexplains the process of colonization that turns barren earth into a living dryland. The pioneers are typically cyanobacteria like the ones in her lab, which can account for up to 95% of the biomass in dryland soil. These were the first organisms to produce oxygen on Earth, about 3 billion years ago, and across global drylands, they bind dust with their tendrilous filaments and sticky secretions known as exopolysaccharides. Over years or decades, more complex lichens and mosses grow and eventually weave an enmeshed substrate rich enough to support larger vascular plants and trees.

This miniature ecosystem acts as the porous frontier between atmosphere and earth: anchoring the soil, drawing down carbon and nitrogen, and absorbing water. Like the fungal networks in forests, these crusts regulate scarce moisture, energy and nutrients via a restless network of interactions between, for example, photosynthetic bacteria that secrete sticky substances, heterotrophic bacteria that consume organic matter and others that fix atmospheric nitrogen by clinging to plant roots. In arid and semi-arid environments, rainfall replenishes no more than half of the water that can be lost to evaporation — but every organism here has evolved to thrive in such conditions, explains Conde Pueyo, including plants with corrugated leaves or layers of hairs adapted to trap cool morning dew. She shows me a red succulent known in Catalan as herba gelada or “frozen grass,” which is covered in glistening, water-filled bladder cells that store water. Where shrubs and trees arise in drylands, they act as vital islands of fertility, producing a microclimate of shade and moisture, and raining down leaves and nutrient-rich detritus. Powerful positive feedback loops can then begin, with soils growing from crusts a couple of millimeters thick into the hefty clod she digs out of the shade, a sample that drips with tendrils dangling down like loose springs.

Southeastern Spain has already been hard hit by climate change, facing heatwaves and prolonged droughts, followed by the types of flash floods that killed hundreds in 2024. The Spanish government has recently backed research to understand the extent of the impacts, finding that 40% of Spain is following Alicante’s path to desertification. The stakes are high planet-wide. Drylands cover most of Africa, the Middle East and Australia, as well as vast swathes of Asia and the U.S. Southwest. Collectively, they accommodate more than 2 billion people, up to 44% of the world’s agriculture, and around one-third of the global biodiversity hotspots. 

The problem is that it’s near impossible to tell from a patch of arid land whether it is healthy or already in decline, Conde Pueyoexplains. Much of what we now know about dryland degradation has been learned since the turn of the millennium, thanks to a small group of ecologists, including Fernando Maestre, who was born a 40-minute drive from Alicante. In 2008, Maestre launched the BIOCOM research project that began setting up open-topped Plexiglas greenhouses across central and southeastern Spain, which raise the temperature by about 2 degrees Celsius and, in effect, provide a portal to the drylands’ near future. The results were alarming. Drylands initially withstood the increased aridity with little impact, then collapsed suddenly. Over email, Maestre recalls looking on in horror as the lichen-splashed biocrusts were “disappearing in front of my eyes.”

“The idea of applying a sci-fi mission of transformation to our own planet is inevitably controversial, and has invited accusations of hubris and ‘playing God.'”

Maestre has built a global dryland monitoring network that searches for early warning signs in soil composition, helped by the development of tools like microbial DNA analysis. In 2020, this data revealed that, as once-humid areas across the planet get drier, they each pass three sudden, catastrophic tipping points. Initially, plants grow smaller leaves to limit water loss, but eventually this strategy reaches a physiological limit, and trees stop growing for much of the year; dryland forests become denuded, leaving behind a landscape resembling the North African steppe. As things get drier still, the soil itself crosses a key threshold, losing microbes like mycorrhizal fungi that cycle nutrients and becoming more vulnerable to erosion. Beyond this, only a few hardy shrubs can survive, thanks to deep roots sunk into the earth. But when aridity levels reach the “ecosystem breakdown phase” — when rainfall replenishes only about one-fifth of what evaporation removes — many of these also perish. The transition to desert is complete.

As climate changes and droughts deepen, half of the planet’s drylands — or 20% of the land on Earth — are predicted to pass at least one of these thresholds by 2100. Worst of all, these changes don’t reverse when the environment cools and rains arrive, leaving the whole ecosystem stuck in its new, depauperate, desertified state. As the planet warms, drylands worldwide are shifting into new equilibrium states, with less life, lower diversity and poorer functioning. This can mean devastation for local communities and wildlife, but the impacts are also felt globally. Collectively, drylands store 46% of the world’s terrestrial carbon, and as these carbon sinks dry, they become net emitters. “So,” Solé asks rhetorically, “what can be done?”

Terraforming Earth

For about a century, scientists and sci-fi authors have imagined what it would take to make another planet’s environment habitable. Jack Williamson, a military meteorologist-turned-writer, coined the term “terraforming” in a 1942 short story, and it has been taken on by the likes of Ray Bradbury, Arthur C. Clarke and Kim Stanley Robinson, whose “Red Mars” meticulously details the transformation of the Martian atmosphere, kickstarted by genetically modified lichens, mosses and algae that are seeded into the inhospitable soil. During the Cold War, scientists, including Carl Sagan and James Lovelock, tore the term from the pages of fiction and applied it to the real world, considering what it might take to turn another planet livable, as nuclear apocalypse threatened to end our own.

The idea of applying a sci-fi mission of transformation to our own planet is inevitably controversial, and has invited accusations of hubris and “playing God.”  Unintended consequences, we’re told, are sure to follow. But Solé has embraced the term. Speaking in 2021 to colleagues at the Santa Fe Institute, a private research center in New Mexico for complex systems theorists, Solé explained that humankind can no longer expect that our surroundings will change in predictable and controllable ways as the world warms by “3-4-5 degrees” Celsius. Instead, our improving climate models show that the coming decades will be defined by the kinds of abrupt tipping points glimpsed in drylands. In the last year, as we reached 1.4 degrees Celsius of global warming, climate scientists declared that Earth had passed the first such planetary inflection point, as warm-water coral reefs began to collapse worldwide. Ocean heatwaves that push heat-sensitive corals beyond their limits can cause entire ecosystems to disintegrate, leading to runaway impacts on the one-quarter of fish species that live on reefs and the one billion people who depend on them for sustenance.

Marine biologists are working hard to save reefs, from coral IVF to artificial clouds, to assisted evolution via selective breeding and synthetic biology. But, as Maestre’s research on irreversible tipping points shows, we are far better placed to save ecosystems before they enter this death spiral. This is the heart of the process that Solé calls terraforming. Sci-fi books’ visions of dramatic, energy-intensive processes — perhaps involving nuclear explosions, orbiting solar sails and clouds of engineered aerosols — won’t be necessary to terraform Earth, he says. Indeed, at the start of his efforts to pursue synthetic biology solutions in 2015, he criticized various “grandiose” geoengineering proposals that required expensive and risky technologies, such as mirrors in space or fleets of aerosol-spewing airplanes. Why construct these machines to fix the environment, he wrote, when nature has evolved self-sustaining, self-replicating living machines? Solé’s approach treats ecological systems as its technological circuitry, taking advantage of the fact that powerful engineers are already in action in every ecosystem. This means we can simply extract one native species and improve it: “So we engineer the ecosystem engineers — that’s kind of the trick,” says Solé.

“If successful, their goal of a single-dose probiotic for drylands, whether gene-edited or hardened, could have astonishing impacts.”

Getting cyanobacteria to sequester more water is perhaps the most straightforward part. Photosynthesizing strains of bacteria like M. vaginatus naturally produce sticky substances in soil crusts as they bind to the soil’s dusty particles. These slime-like substances retain moisture from mist and rain, making it available to other organisms. Synthetic biologists have been editing bacteria for years to produce hyaluronic goo and other substances that super-charge this process, explains Victor Maull, a 20-something specialist in synthetic ecosystems who’s a postdoc in Solé’s lab. Maull has developed “genetic circuits” to get bacteria to produce hyaluronic acid, which are now ready to be tested in a model organism called Bacillus subtilis.

A small change in moisture levels can make a big difference. In Alicante, wherever a little more water could be retained — in tree-shaded patches or depressions in the soil — that was enough to kickstart a feedback loop that enabled ever more complex biocrusts to form. “This is multiplicative,” Solé says. He calls it a synthetic mutualistic loop: If it works, “the hope is our synthetic strain will improve soil quality; soil quality improves plant cover; and plant cover improves soil quality.”

So far they’re reliant on computer models because the 1980s backlash over GMOs had a chilling effect on research, Solé explains, meaning we know frighteningly little about how synthetic microbes interact with wider ecosystems. “It’s terra incognita,” he says. But the real-world data is now close at hand. 

The Mesocosm

At Barcelona Biomedical Research Park, Solé’s lab sits one floor below the European Molecular Biology Laboratory (EMBL), an eminent 50-year-old institution that led Europe’s part in decoding the human genome. Conde Pueyo has spent her time between the two — originally working on Solé’s team when the project began, before transferring the short distance to the medical-grade research facilities where the mutant microbes have been growing. Hidden behind a barricade of furniture to ensure it remained intact, Conde Pueyo has built a replica of the Alicante dryland inside a climate-controlled chamber about the size of a fridge-freezer. It has taken roughly two years since Solé first told me of his plans for a realistic “mesocosm” — an enclosed experimental ecosystem—to accurately simulate such an arid environment. The project, named Synterra, is likely the most sophisticated replica to date and is led by EMBL, which has, in recent years, taken a growing interest in microbiology beyond the human body to understand how life is responding to a changing planet.

The developmental biologist James Sharpe, who leads EMBL Barcelona, encouraged her to construct the soil system to the exacting standards of a human “organoids,” three-dimensional tissues grown from stem cells, which in the last decade or so have revolutionized how her EMBL colleagues research chemotherapy drugs and human development. Organoids are a “halfway house” between experiments on individual cells and full test subjects, explains Sharpe. In effect, the Synterra project is constructing an “ecosystem in vitro,” Sharpe says, to help researchers determine whether drylands are reaching tipping points and test out potential solutions.

Today, the mesocosm traps the soil microbes from Alicante in a “Truman Show”–style simulation, recreating daily atmospheric cycles over the course of a year. Just as the drylands chill overnight and gradually rise to become sunbaked at the height of the day, so would the dryland inside the mesocosm, with precisely matched humidity, rain, temperature and UV exposure. There is a limit to how perfect this simulation can be, she explains: inside, there’s no room for rabbits scattering droppings or ants that carry microbes around. Likewise, the simulation runs one month and 12 hours behind the outside world. At the start of March, Conde Pueyo programs the mesocosm’s climate control system using February weather data from Alicante, allowing it to reproduce the previous month’s temperature and precipitation. The additional half-day delay allows her to repair any broken lights during the microbes’ “nighttime.” Other than that, the dryland lives out the same year its wild partners do.

Greening The Dryland

The team is now fine-tuning this simulation to ensure it is a reliable twin of the real world, which will allow them to accurately explore its possible futures. Ultimately, the mesocosm could provide the necessary data to show if these mutant microbes will explode in number, causing cascades of extinctions of other organisms in the dryland crusts. All the Complex Systems Lab’s computer models so far have shown that the synthetic bacteria will be safely controlled without the need for an engineered kill switch. Microbes could be programmed to work only in narrowly defined conditions — like a certain level of dryness — and die when they’re done rehydrating the soil, explains Solé, who holds advanced degrees in biology and physics.

“In environments approaching tipping points, it’s clear we must do something or face certain collapse.”

But Solé has argued that diverse ecosystems, like those found in soil crusts, will naturally limit the impact of this new arrival, just as a bunny rabbit dropped off in the Amazon or the Everglades stands little chance of taking over. Invasive species are the rare exceptions, defined as introduced organisms that cause significant negative impacts to native ecosystems and human societies. These invasives — like the highly toxic cane toads that have spread across Australia, poisoning predators — cause massive ecological harm. But 99% of introduced organisms don’t become a runaway problem. “Biodiversity is a strong firewall that can contain and control things,” he says.  

In fact, the hardening process, by identifying the most resilient bacteria and breeding them in large numbers, should ensure that the bacteria’s spread is not overlycontrolled. This process reveals bacteria with genes conferring tolerance to heat and aridity — adaptations that will help address climate challenges in the years to come. But Conde Pueyo believes the hardened, surviving microbes could be a shovel-ready solution. By growing billions of these pre-adapted ecosystem engineers, they have already created a potential dryland probiotic — one that can nudge this ecosystem toward resilience, like how heat-trained corals are being planted in tropical reefs. Rather than waiting for a species to evolve resistance to heat or aridity, you instead “take species from the future and you bring it back to now”, she says, and see if the whole community benefits. This treatment is already being tested in the mesocosm, where, over the next year, her initial experiment will test patches of dryland — “soil organoids” — that have been dosed with a mixture of hardened bacteria, against those that have not.

If successful, their goal of a single-dose probiotic for drylands, whether gene-edited or hardened, could have astonishing impacts. It’s difficult to say how much carbon could be captured if drylands are protected and restored, but some soil scientists have estimated it could be as much as a gigaton of carbon per year, roughly equal to the amount nations have committed to capture under the Paris Agreement.

Yet dryland restoration remains in its infancy. Most efforts so far have been adapted from reforestation practices in attempts to erect basic barriers to desert expansion, such as the African Great Green Wall. This ecological megaproject began in 2007 with a pledge by the African Union to build a 10-mile-wide belt of trees along the 5,000-mile Sahelian drylands. But nearly two decades later, there has been little progress. Millions of trees were planted, but in some areas an estimated 80% of them have died. 

After decades of tree-planting failures, global plans to restore drylands are being redrawn. China’s own green wall project in the Tengger Desert, a 95-mile-long and 6-mile-wide wall along a railway line, initially experienced similar mass mortality, but over 30 years, Chinese restorationists developed a “checkerboard” technique that half-buries straw in a grid-like pattern to anchor the sand and deposits organic material like manure in the middle to help plants establish. Increasingly, everyone realizes you must start with the soil, says Sonia Chamizo, an ecologist at the University of Almeria in Spain’s ultra-dry southeastern corner. Chamizo is part of a team funded by the Spanish Science Ministry, one of only a half-dozen labs researching methods for biocrust-driven restoration, along with teams in China, Australia and the U.S. The usual method involves extracting small pieces of biocrusts from healthy areas and growing them in greenhouse environments, before transplanting them into degraded areas. This type of “soil skin graft” has been pioneered by the world’s first-ever “biocrust farm” near Moab, Utah, established to restore the jaw-dropping arid landscapes where three-dimensional crusts have been compared to fried eggs, shag carpets and extremely toasted marshmallows.

Teams in the U.S. and Europe believe one key to upscaling their operations might be solar power plants, which are rapidly expanding across global drylands and creating new microclimates of tree-like shifting shade. “We want to use these photovoltaic farms as nurseries to grow biocrust and plants,” says Chamizo. Picture rows of solar panels with a microscopic rainforest sprawling out beneath. At a pilot in Almeria, the team is planning to restore the biocrust by cultivating bacteria and mosses on site and by “biopriming,” or dipping plant seeds in microbes to help them grow roots in dry soils. They’re still at an early stage, she says. But the team has devised lower-cost chemicals to feed the bacteria and is trialing new additives to boost effectiveness, including compost made from paper factory waste that helps bind bacteria to the soil. In time, they imagine pumping out the foundational ingredients — a slurry of seeds, mycorrhizae and fertilizing chemicals — via hydroseeding techniques. This would effectively spray a ready-made dryland out of a hose onto degraded soil.

“Would we ever look at the African savanna the same way if we knew that tiny synthetic mutants were possibly out there spreading cosmetic goo through the soil? Does this so-called natural ecosystem become a lie?”

Synthetic Ecosystems

Amid today’s restoration experiments, Solé and his team argue that gene editing is not only a promising approach but also a comparatively modest one. “It might not look like it, but it’s more conservative compared to what has been proposed,” the ecologist Maull tells me. In environments approaching tipping points, it’s clear we must do something or face certain collapse. Reefs show us the way. Roughly 90% face extinction by 2050, leaving environmentalists scrabbling for answers and trialing powerful technologies for conservation and repair, including gene editing, small-scale geoengineering and countless other odd inventions.

Kent Redford, who ran a task force on synthetic biology for the International Union for Conservation of Nature (IUCN), has argued that today’s environmental crises make clear that conservation needs to add to its toolbox of solutions. This includes understanding the risks and benefits of genetic engineering. “I have had arguments, particularly with Western environmentalists, [who say] this should never be even considered,” Redford, a conservation biologist and co-author of the 2021 book “Strange Natures: Conservation in the Era of Synthetic Biology,” tells me. Solé’s work is bound to face the kinds of charges already leveled against gene drives, a potential solution to the wave of rare reptiles and birds driven to extinction by rodents that arrive on small islands from ships. Gene drives use CRISPR to modify sections of genetic code designed to be passed down to offspring. In this case, the code would be designed to spread female infertility traits through wild rat or mouse populations. Releasing a small number of genetically modified adults onto an island would eventually cause the invasive population to collapse. “I think it’s unconscionable not to give serious consideration to them,” he says.

Critics have warned that, if rats escaped the island, the gene drive could go global, killing off rodents across the planet. “If we’re going to find solutions that work, we’re going to need to have them spread on their own,” Redford says. “And that need to have self-propagating solutions is even scarier to people.” Like rats, microbial solutions rely on being able to reproduce in the wild but are even less containable. If they don’t end up being neat solutions but problems, then you’ve got a self-perpetuating crisis on your hands. Just as alarming, say critics, in the microbial world, DNA is also commonly transferred across species boundaries in a process of horizontal gene transfer, meaning entirely unrelated species could end up producing hyaluronic goo.

Solé is clear that, for now, they’re testing and learning whether these unintended consequences actually occur. In work with microbiologist Victor de Lorenzo, he has argued that horizontal gene transfer can be an asset, allowing them to introduce the goo-producing trait without hacking the bacteria’s core chromosomal DNA; instead, introducing it as a supplementary strip of DNA known as a plasmid that the bacteria may eject or transfer in time. Solé asks what the public would prefer: one gene-tweaked microbe in the soil, or something like China’s labor-heavy approach in the Tengger Desert? “To do that, you bring tons and tons and tons of dryland and haystacks from somewhere else, which means full microbiomes that are totally exotic. Why is that not a problem?” he asks. But “a single gene, put in a cyanobacteria that was [already] in the community, is a problem?”

It’s the kind of once-unthinkable argument that synthetic biologists are starting to win. Last October, Conde Pueyo and Solé held their breath as environmentalists debated a blanket ban on the release of genetically modified organisms into the wild. A key vote in Abu Dhabi at the IUCN World Conservation Congress, the world’s biggest meeting of conservationists, could have forced them to shut down the mesocosm just as it was finally up and running. But to the surprise of many, more than 250 scientists signed a letter highlighting the wide-reaching potential of synthetic biology, from eliminating invasive rodents to protecting horseshoe crabs from overexploitation by developing synthetic blood. The final tally was split nearly perfectly, and the ban failed by a single vote, paving the way toward the release of synthetic organisms into the wild. In the decade that Solé has traveled to speak on this, synthetic biology has been embraced by scientists and restoration practitioners working in reefs, forests and the Arctic tundra. “We are not, anymore, these crazy guys saying crazy things,” he says.

“Most of us continue to believe that we can continue to live in the relatively stable world we’ve lived in for 10,000 years or more, but this is a kind of ‘holocene hallucination.’”

Even if the mesocosm experiments end up showing that the gene-edited microbes are safe and effective, there remains other, perhaps deeper, worries. Many in the public still want their natural systems unmolested and engineered microbes plainly are something else. Would we ever look at the African savanna the same way if we knew that tiny synthetic mutants were possibly out there spreading cosmetic goo through the soil? Does this so-called natural ecosystem become a lie?

Redford says it would be a mistake to simply wave our hands and think we can forbid anything “unnatural.” Organisms are already being edited in agriculture and biomedical sciences, with foreseeable spillover into wild environments. “I see that objection from a set of people who are — it’s a kind of pejorative term — backwards looking,” he says.

Most of us continue to believe that we can continue to live in the relatively stable world we’ve lived in for 10,000 years or more, but this is a kind of “holocene hallucination,” he says. “And the reason it’s a hallucination is that it flies in the face of all evidence. We can’t have that world back.” In the Anthropocene, as drylands hurtle toward climate-induced tipping points, pristine nature is in the rear-view mirror. Already, nature reserves are forced to reckon with the fact that climate change and pollution are pushing their landscapes into wholly new states, dubbed “novel ecosystems,” with forecasts showing that such historically unprecedented ecosystems could make up more than 50% of terrestrial ecosystems by 2100. Within fields like restoration and rewilding, these freakish ecosystems are increasingly being seen as the Anthropocene’s new normal, worthy of scientific attention and conservationists’ care. As the split-vote in Abu Dhabi made clear, there is a growing tension between the priorities of conventional conservation and “people like us,” says Maull, who embrace these new futures. 

Green Mountain Glory

There are plenty of proposals to genetically modify organisms to suck up carbon, like Bay Area startup Living Carbon, which has gene-edited poplar trees to grow and absorb CO2 rapidly. Likewise, the Dallas, Texas-based “deextinction” startup Colossal Biosciences touts their synthetic mammoth as helping to lock up greenhouse gases in the Arctic permafrost. What sets Solé’s microbial solution apart is their potential impact at a planetary scale, says Redford. “De-extinction, and all of that sort of thing, makes absolutely no difference in the future of the world. It doesn’t matter at all,” says Redford. By contrast, Redford notes, an effective remedy for dryland degradation could make “a huge difference in how habitable and where habitability is going to be possible in this world.” Solé is convinced that a ready-made microbial solution would win over doubters, pointing to genetically modified mosquitoes already released in several countries that have become an effective precaution against malaria.

A fortnight after I’ve come home from Barcelona, Maull messages me on WhatsApp to say he has finally made it via military plane to a corner of the world that demonstrates how dramatic the impact could be. Ascension Island is a volcanic rock in the South Atlantic, claimed by the British Crown in the 19th Century as a base to guard Saint Helena, where Napoleon Bonaparte was imprisoned. Back then, Ascension was little more than a mound of black basalt with a smattering of ferns, what Charles Darwin called a hideous “cinder” in 1836. Yet thanks to Darwin and his friend Joseph Hooker, the island has a peculiar place in ecological history. For three decades, the British transported hundreds of species from across the planet to Ascension: hardy thorns, wild tobacco plants and periwinkles, along with donkeys, rabbits and bees. Maull was thrilled to explore the wholly man-made cloud forest that had assembled itself out of this chaotic legacy. At the summit of Ascension’s Green Mountain, he walked through the very “densest jungle” where banana trees, bamboo, yew and juniper were braided into thick cosmopolitan canopies. 

For the terraforming project, pushing back today’s tipping points is the primary target, but not the only one they have in mind. Maull has flown here to determine whether the island can provide engineering lessons for what are known as positive tipping points, in which ecosystems flip to a more verdant, self-sustaining state. Although far messier than the lab’s clinical experiments, Ascension’s transformation relied on a similarly simple effort to trap water, as Hooker set out to erect foliage to capture the moisture that would otherwise pass by on the ocean breeze. Over the course of nearly two centuries, this unruly experiment proved successful when measured by biomass and biodiversity, which have boomed, providing habitats where native crabs and feral donkeys can multiply.

“As drylands hurtle toward climate-induced tipping points, pristine nature is in the rear-view mirror. Already, nature reserves are forced to reckon with the fact that climate change and pollution are pushing their landscapes into wholly new states.”

It now seems inevitable that many more ecosystems will join coral reefs and degraded drylands in crossing tipping points before humanity gets its act together and makes successful interventions. Already, 12 million hectares of drylands are degraded every year, part of a restless global shuffle, with humid areas becoming drylands, then deserts, as other areas move in unpredictable directions. With global temperatures rising, in some places we will inevitably cross irreversible thresholds. Conservation often finds itself trapped trying to preserve the past — like the Great Green Wall’s trees, planted into a landscape that can no longer support them — but Ascension shows us there may be a choice, says Conde Pueyo. Ecosystems have the capacity to reform and reconstitute in ways we might never have imagined. When the past cannot be maintained or recreated, communities in the Sahel might be forced to act like those in Ascension or Mars. “They need to reinvent a future: a new ecosystem for the new harsh climate,” Conde Pueyo says.“This romanticism about just going back to the old steady state, maybe it needs to be abandoned.”

The lab is working to ensure the “systemic properties” of a healthy ecosystem, such as biodiversity, which helps an environment remain resilient and self-sustaining, Maull says. To him, the actors in this biodiversity are less important. Tomorrow’s thriving ecosystems could be built from organisms that — like the European bees pollinating Himalayan flowers on Ascension — evolved in entirely different parts of the world. Or, like the lab’s synthetic species, they are not simply the product of natural evolution. For Maull, Ascension is “a way out. An old scientific experiment that proves ecosystems can be designed,” he messages me in WhatsApp. “Ecology was not even a scientific discipline when the cloud forest in this place was invented. We now have better tools and wider knowledge to rethink these questions.”

It’s tough for people to face the fact that the Holocene Eden is gone, and we must now imagine a different world, explains Redford. But he’s optimistic they can make the switch. He recounts how, as recently as the 17th century, travelers crossing the Alps drew the curtains of their carriages to avoid gazing at the mountains. In “Mountain Gloom and Mountain Glory,” literary scholar Marjorie Hope Nicolson tells the story of how they saw these ragged peaks as warts on a world created as an ordered and gentle garden. In a century or two, the mountain-loathing perspective was lost entirely in a shift driven by Romantic writers like Mary Shelley, who heralded the experience of the mountain as an encounter with the transcendent. 

“I think we’re in a kind of ‘Mountain Gloom, Mountain Glory’ phase now,” Redford says. We’re still averting our eyes from a world filled with forbidden tools like gene-editing and impure landscapes like the Green Mountain. “We might as well push the curtains back and have a look out there and figure out how we want to live in that sort of a world.”