🎧 Prefer to listen? I've recorded this essay for you — press play above.

For nearly two decades, the UK presented itself as a climate leader. The 2008 Climate Change Act created the world’s first legally binding national carbon budgets; in 2019 Theresa May’s government made the UK the first major economy to legislate for net zero by 2050; and successive governments treated falling territorial emissions as evidence that decarbonisation and economic prosperity could proceed together.

UK greenhouse-gas emissions are now about 50% below 1990 levels, reaching 407 MtCO₂e in 2025. But the political confidence surrounding that project has become much less secure.

Russia’s invasion of Ukraine exposed the vulnerability of energy systems to gas prices. UK and European industrial gas prices tripled between 2020 and 2022 and remain more than twice their 2020 level, while several energy-intensive UK manufacturing sites, including Grangemouth in Scotland, closed in 2025. Industrial electricity also remains more than four times the price of gas, complicating the very electrification on which decarbonisation depends.

At the same time, another government priority is demanding remarkable amounts of electricity. The UK wants to become an AI leader and is creating AI Growth Zones around data centres, explicitly seeking sites with 100 MW-plus grid connections and proposals for clusters reaching at least 500 MW by 2030. The government’s own strategy now describes domestic computing capacity as critically necessary for economic growth, resilience and the protection of sensitive data.

These are no longer marginal tensions around an otherwise settled net-zero pathway. Climate security, energy affordability, industrial production and digital sovereignty are becoming competing claims on the same electricity system. The question is shifting from how quickly can emissions be reduced? to something more difficult, which futures are being secured, for whom, and what happens when they cannot all be built at once?

The Future Without a Past

There is something peculiar about the word transition. It suggests movement from one condition to another. One system recedes and another takes its place. Coal gives way to oil, fossil fuels to renewables, the high-carbon past to the low-carbon future. This is now so embedded in climate policy that it can seem like a description of how energy systems naturally change.

But the history of the idea is much more complicated. As the energy historian Jean-Baptiste Fressoz argues, the modern concept of energy transition emerged partly through futurology, attempts, from the mid-twentieth century onwards, to calculate how much energy societies would need in the future and which sources would provide it. From this came the familiar sequence of wood → coal → oil → gas → renewables, with one dominant source apparently replacing another. The problem, Fressoz argues, is that this picture of the future was then used to tell the story of the past. Historically, energy sources rarely disappeared when new ones arrived; they accumulated, interacted and often enabled one another. In other words, an imagined future helped produce a distorted version of the past.

Look more closely and the sequence becomes much messier. Coal did not simply replace wood. Oil did not make coal disappear. Nuclear power and natural gas were added to energy systems already consuming enormous quantities of fossil fuel. As his book puts it, energy sources have historically been “as much symbiotic as they are in competition”. They have tended to add up rather than substitute each other.

This matters because the same image of succession now structures the future being built around net zero. Install enough wind and solar, electrify enough cars and homes, and eventually fossil fuels move out of the picture. The pathway bends downwards.

But what if transition gives us the wrong picture of what is happening?

Perhaps the difficulty is not only that the energy transition is moving too slowly. Perhaps new energy systems have never arrived on an empty stage.

Addition, Not Substitution

The distinction is important because the atmosphere does not care what percentage of an energy system is renewable. What matters for climate change is how much carbon continues to accumulate.

Renewables can therefore grow exceptionally quickly without producing an equivalent decline in fossil energy if total energy demand is growing at the same time. Electrification expands the work we expect electricity to perform. Cars, heating and industrial processes all move onto the grid. Artificial intelligence (AI) and data centres add another source of demand. Hydrogen production and carbon removal could add others. Each may have a defensible claim to electricity, while together they produce an increasingly large energy system.

China makes this visible at extraordinary scale. It is building more solar and wind than any other country while continuing to maintain and expand a vast coal-power system. This is often presented as a contradiction, how can a country be leading the renewable transition while still building coal? But perhaps China is showing us something historically familiar. New sources of energy do not necessarily remove old ones. They can accumulate alongside them and even give them new functions. Coal that once provided continuous baseload electricity can also become capacity held for periods of high demand, low renewable generation or disruption elsewhere in the system.

The problem is that climate change ultimately requires something much more difficult than additionality. However many solar panels, wind turbines, batteries or carbon-capture plants are constructed, absolute fossil-carbon emissions still have to fall.

There is an disturbing asymmetry here. Modern political economies are extremely good at addition. A shortage of electricity calls for more generation. Congested grids call for more transmission. Intermittency calls for more storage. Industrial competition calls for more manufacturing. AI leadership calls for more computing, which calls for more data centres, electricity and grid capacity.

Climate physics, meanwhile, requires subtraction.

So perhaps one of the more difficult questions for transition is not simply what needs to be built? It is what are we prepared to stop?

Living Well With Others Within Limits

This is where Julia Steinberger’s Living Well Within Limits project becomes useful to me. Instead of beginning with projected energy demand and asking how enough low-carbon supply can be built to meet it, the project asks a different question, what resources are actually necessary for people to live well?

Energy, after all, is not something most people want for its own sake. What matters is what it makes possible. Warmth, food, mobility, healthcare, education, relationships, security and time. But access to those things—and the energy used to provide them—is profoundly unequal. In 2019, the highest-income 10% of UK households—measured by household income adjusted for household size—used nearly three times as much energy as the lowest-income 10%. The richest tenth used more energy flying alone than the poorest tenth used across everything they consumed.

That makes the apparently neutral question how much energy will the UK need? much harder. Need for what, and for whom? Heating a cold home is not the same demand as another long-haul flight. A hospital and a hyperscale AI data centre might both appear in an energy model as demand, but they do not make equivalent claims on a shared future.

For the last couple of years, living well with others within limits has become something of a maxim for me. Living well asks what energy is actually for. With others extends that question to the people, landscapes, species and future generations implicated in producing what is consumed here. And within limits asks something contemporary politics finds much harder. What happens when not every demand can continue expanding?

Limits, then, are not simply about consuming less. They raise questions of distribution, sufficiency and excess. Who has too little, who consumes disproportionately, and which demands are being treated as necessary? Net zero, reindustrialisation, heating, transport, AI and digital sovereignty can all make claims on the same electricity system. But once there are limits, necessity becomes political.

Which leads to another question. What, exactly, is being secured?

What Is Being Secured?

Security has moved much closer to the centre of energy politics. Russia’s invasion of Ukraine made this explicit. European gas prices had already been rising, but the invasion and Russia’s restriction of supplies exposed what dependence on an internationally traded fuel could mean for households, industry and governments. The UK’s 2022 Energy Security Strategy responded by calling for more ”homegrown” energy and greater energy independence; a year later the government created the Department for Energy Security and Net Zero (DESNZ), placing the two objectives in the same institutional name.

But energy is only one of the things governments now want to secure. The USA has treated access to advanced semiconductors and AI infrastructure increasingly as a matter of national security. China’s dominance of solar panels, batteries and electric vehicles (EVs) has made manufacturing capacity geopolitical. And the UK is now using strikingly similar language around computation. Its 2026 Compute Roadmap describes computing not simply as infrastructure but as a source of “sovereign capability” — the national-security idea that a state should retain the capacity to act without depending on others; the government wants domestic capacity stretching, in its ideal formulation, from chip to system to software. Up to £2 billion is being committed to the national compute ecosystem, alongside a £1.1 billion AI Hardware Plan announced in June 2026.

There is a recognisable logic here. A country dependent on another for its gas, semiconductors, cloud computing, batteries or electricity has fewer choices when trade fractures, war begins or supply chains fail. But sovereignty introduces another future into the climate calculation. The question is not just how to produce electricity with fewer emissions. It is which infrastructures must exist domestically so that a country retains the capacity to act.

And this brings me back to the previous question. Security sounds self-evidently desirable until its object is specified. Security of energy supply? Industrial production? Data? Military capability? GDP? Cheap consumption? A particular standard of living?

What, exactly, is being secured — and for whom?

The Computational Future Has Already Arrived

I ended my last essay with computation because it kept appearing inside supposedly non-computational infrastructures. A wind turbine can operate without AI, but contemporary renewable systems increasingly depend upon forecasting, sensors, optimisation, automated maintenance and the computational systems needed to coordinate electricity across increasingly complex grids.

The UK is now planning for this dependency at another scale. Its AI Growth Zones are explicitly intended to accelerate large data centres by overcoming two of their biggest obstacles, planning and access to power. The government says its reforms could cut “time to power” by as much as five years and save a 500 MW data centre up to £80 million annually in electricity costs.

That phrase — time to power, the total time a data centre needs to get a working electricity supply from the grid — is revealing. AI is often discussed as software, intelligence, information or productivity. But making AI sovereign means making it physical. Data centres need land, substations, transmission, cooling, chips and enormous quantities of electricity. Chips require fabs, ultrapure water, minerals and globally distributed manufacturing systems. The cloud eventually comes back to earth.

And its carbon consequences are not trivial. The UK government’s own analysis estimates that cumulative emissions from AI compute could reach 34–123 MtCO₂ over the coming decade, depending upon demand, efficiency and the rate at which the electricity system decarbonises.

This is another form of additionality. The computational revolution did not replace the industrial revolution. It inherited it. Steel, concrete, copper, silicon, electricity and water have been reorganised around another set of promised futures.

The interesting question for climate politics is not whether AI can become more efficient or run on renewable electricity. It is what happens when a future built around rapidly expanding computational capacity encounters an energy transition that is itself trying to electrify transport, heating and industry.

Both futures require electricity.

Both require infrastructure.

Both describe themselves as necessary.

The Things Put Somewhere Else

There is another reason this materiality can be difficult to see from the UK.

The UK’s territorial emissions have fallen dramatically. Provisional figures put them at 367 MtCO₂e in 2025, 54% below 1990 levels. This is real progress, particularly the extraordinary removal of coal from electricity generation. But territorial accounting ends at the national boundary. It does not include the emissions produced elsewhere making the solar panels, computers, clothes, cars and other goods consumed here.

Measure the UK through consumption instead and a different landscape appears. The latest footprint figures put UK consumption emissions at 699 MtCO₂e in 2023. More than half of that footprint came from imported emissions. Since 1996, territorial emissions have fallen by around 50%, while the consumption footprint has fallen by only 15%; emissions associated with imports actually increased by 43%. China alone accounted for 13% of the UK’s entire consumption footprint in 2023.

This doesn’t make the UK’s emissions reductions imaginary. It shows how much depends upon where the boundary is drawn.

Manufacturing can move somewhere else. Mineral extraction can happen somewhere else. A data centre can run somewhere else. Carbon can consequently disappear from one national account without disappearing from the atmosphere.

But sovereignty is beginning to push in the opposite direction.

Semiconductor security says some chip capacity should come closer. Energy security says more electricity should be generated domestically. Digital sovereignty says sensitive computation and data should not depend entirely upon infrastructure overseas. Industrial strategy talks again about manufacturing things in the UK.

There is something deeply unsettling about this reversal. For decades, globalisation allowed wealthy economies to separate consumption from many of the landscapes, infrastructures and emissions that made it possible. Extraction happened elsewhere. Manufacturing happened elsewhere. Increasingly, computation could happen elsewhere too. Now some of those distances have begun to look dangerous. Energy dependence is vulnerability. Overseas manufacturing is supply-chain risk. Foreign computing is a question of digital sovereignty. And so, governments are beginning to bring certain infrastructures back within view. Not because inequity became intolerable. Not because environmental injustice finally demanded it. Because dependence became a security problem.

At the same time, a cloud service running elsewhere remains almost immaterial. A 500 MW data centre proposed nearby becomes land, pylons, electricity, heat, water and planning.

Perhaps sovereignty does not only bring production home. It brings some of its material consequences home too.

Energy Inheritances

This is where I return to the problem with climate transition.

If energy history is additional rather than simply substitutive, perhaps the infrastructures now being constructed should not only be understood as replacements for what came before. They are also inheritances.

Renewable electricity inherits transmission systems, mines, ports, financial institutions and industrial capacities built through fossil energy. AI inherits electricity grids, semiconductor industries, data networks and enormous systems of mineral extraction and manufacturing. Energy security inherits memories of shortages and geopolitical shocks. Digital sovereignty inherits a world in which technological capacity has become concentrated in a number of corporations, countries and supply chains.

None of them begin from nothing. And each produces inheritances of its own.

The wind farms, transmission lines, batteries, data centres, semiconductor plants and AI systems being constructed now will shape what future governments are able to do. They create skills and capacities, but also maintenance requirements, resource demands, sunk investments and dependencies. Infrastructure does not simply enable a future. It makes some futures easier to inhabit and others harder to leave.

This is what I mean by an energy inheritance. It is related to what energy researchers call lock-in, the way a system, once built, becomes difficult to leave behind. Build a gas power station and it is not only the physical plant that persists. Money has been invested on the assumption that it will operate for decades; workers and supply chains form around it; regulations and markets adapt to it; pipelines connect to it; homes and industries come to depend on the energy it provides. Closing it therefore means undoing a whole set of relationships, investments and expectations, not simply switching off a machine. Infrastructure, in this sense, helps make its own continuation seem necessary. Inheritance asks something slightly different. Rather than only asking why the old system persists, it asks what the supposedly new system receives from it. Renewable energy inherits grids, mines, ports, engineering knowledge and industrial capacity built through fossil energy. The computational future inherits electricity systems, mineral extraction and semiconductor manufacturing. It may also inherit fewer tangible things. Assumptions about endless energy availability, expectations of economic growth, inequalities between places and particular ideas about what prosperity or security should mean. Lock-in asks why it is difficult to leave a world. Inheritance asks how much of that world comes with us when we think we have left.

Perhaps, then, the problem with imagining an energy transition as movement from one era into another is not simply historical accuracy. It encourages the possibility of a future without a past.

My work on Carbon Afterlives has increasingly made me suspicious of that possibility.

There is no clean break between the industrial and computational worlds, fossil and renewable infrastructures, national prosperity and landscapes elsewhere. Futures arrive already carrying other worlds inside them.

So, the question may not be which of the UK’s competing futures should win, climate leadership, industrial growth, energy sovereignty, AI leadership or affordable everyday life.

The harder task is deciding what is worth securing, what is enough, what must actually be allowed to decline, and what others will inherit from the futures being secured now.

x

Across UK, Europe and North America, opposition to artificial intelligence (AI) increasingly focuses on data centres. They consume enormous quantities of electricity, water and land. They require vast amounts of computation to sustain systems whose social value is often unclear. As climate concerns deepen, data centres have become an unlikely symbol of digital excess; another form of infrastructure whose environmental costs appear increasingly difficult to justify.

At the same time another, quieter, story is unfolding.

Across landscapes and coastlines, wind farms are becoming central to climate transition. They promise electricity without combustion, lower emissions and pathways beyond fossil fuels. At the same time, they are also becoming computational environments. Cameras identify birds before collisions occur. Machine learning models detect faults in turbine gearboxes months before engineers would otherwise discover them. Forecasting systems estimate tomorrow’s electricity generation before the wind has even arrived. Grid optimisation models continually coordinate renewable electricity across increasingly complex energy networks.

These developments rarely appear in the same conversation.

Data centres are discussed as though they belong to one environmental problem, while renewable energy belongs to another. Computation appears on one side of climate politics; decarbonisation on the other. Yet the more time I spend following renewable infrastructures, the more difficult this separation becomes to sustain.

What is striking is that almost nobody asks how much computation should be devoted to something like renewable energy. Public debates tend to treat AI as a single, homogeneous technology whose environmental costs can be aggregated into the electricity demand of data centres. However, the computation used to forecast renewable generation, monitor wildlife or predict turbine failures represents a very different kind of workload from training frontier language models or serving billions of consumer AI queries. We have become accustomed to debating how much energy AI consumes without asking what that computation is actually doing.

One way of asking this question is deceptively simple: could we run a wind farm without AI?

The immediate answer is yes. Wind farms have operated successfully since the 1980s, long before contemporary machine learning became part of renewable energy systems. Engineers maintained turbines, ornithologists monitored bird populations, meteorologists forecast the weather and grid operators balanced electricity systems using statistical models, operational experience and established forms of automation. Renewable energy does not suddenly stop without AI.

But that is perhaps the wrong question.

For some time now, Carbon Afterlives has been following the worlds that gather around climate interventions, the infrastructures, obligations and forms of stewardship that emerge long after a technology is introduced. Carbon stores require monitoring long after emissions have been captured. Hydrogen infrastructures assemble new technical expertise and institutional commitments long before their futures are secure. Climate transitions do not simply replace one system with another; they accumulate new relationships that others will inherit.

AI may be another of these inheritances.

Instead of asking whether AI is good or bad for the energy transition, this essay follows computation through a wind farm. It traces bird monitoring, predictive maintenance, weather forecasting and grid coordination to ask a question, what kinds of computational obligations are quietly becoming embedded within the infrastructures of decarbonisation. If fossil fuels left us pipelines, refineries and atmospheric carbon, what afterlives are being assembled by renewable energy?

Does a Wind Farm Need AI to Protect Wildlife?

The short answer is no.

Long before AI entered the wind industry, protecting biodiversity formed part of the everyday work of operating a wind farm. Before construction began, ecologists mapped migration routes, identified breeding sites and assessed collision risk. Once turbines became operational, ornithologists continued monitoring bird populations, documenting mortality and advising when turbines should be curtailed to protect vulnerable species. Wind farms have never depended on AI to care for wildlife. They have depended on ecological expertise.

So why introduce AI at all?

The answer lies less in replacing human knowledge than in changing how that knowledge is enacted.

Increasingly, ecological monitoring is becoming computational. Systems such as IdentiFlight combine high-resolution stereo cameras with machine learning models to monitor the airspace around wind turbines continuously. They identify protected bird species, estimate flight trajectories and calculate whether a bird is likely to enter the rotor-swept zone. If the predicted risk exceeds a predefined threshold, the system automatically signals the turbine to slow or stop—often within a matter of seconds, before the bird reaches the blades. Unlike periodic ecological surveys or human observation, this process operates continuously, twenty-four hours a day, across an entire wind farm.

This is often presented as an example of AI making renewable energy more environmentally responsible. In one sense, it is. AI enables ecological knowledge to be enacted in real time and at a scale that would be impossible through human observation alone.

However, the technology does not decide which birds should be protected, what level of risk justifies shutting down a turbine, or how ecological value should be balanced against energy production. Those priorities have already been established through environmental legislation, planning conditions, ecological surveys, conservation objectives and regulatory processes. The AI does not determine ecological value; it operationalises priorities that have already been negotiated by scientists, policymakers and regulators. Having said that, this does not mean environmental governance remains unchanged. Decisions about where AI should be deployed, how its recommendations should be used and what role it should play in environmental protection are themselves becoming part of contemporary governance. AI is not external to environmental regulation; it is increasingly being woven into the infrastructure through which environmental care is organised.

Could a wind farm function without this technology?

Almost certainly.

Wind farms operated for decades before computer vision became available, and many continue to rely primarily on ecological surveys, seasonal monitoring and operational curtailment. AI is not a precondition for renewable energy. It is an additional layer of environmental governance.

What changes, however, is the nature of ecological care.

The bird no longer appears only through field observations, binoculars and ecological surveys. It also becomes a sequence of pixels, a classified species, a predicted trajectory and a quantified probability of collision. Ecological intervention is reorganised through computational practices that make some relationships visible, actionable and governable in new ways. Responsibility remains distributed across ecologists, engineers, regulators and operators, but it is increasingly exercised through models that classify, predict and intervene before harm occurs.

Following AI through bird monitoring changes the question. The issue is not whether renewable energy can exist without AI. It can. The more pertinent question is what kinds of environmental care become possible once computation becomes part of the infrastructure through which biodiversity is protected.

Does a Wind Farm Need AI to Maintain Turbines?

Again, the short answer is no.

Wind farms have always depended upon continual maintenance as opposed to AI. Engineers routinely inspect turbines, replace worn components and respond when equipment fails. Bearings wear, gearboxes overheat, blades accumulate damage and electrical systems drift beyond their expected operating conditions. Maintaining a wind farm has always involved recognising these changes before they become catastrophic, drawing upon engineering expertise, routine inspections and operational experience.

So why introduce AI?

As wind farms have become larger, more remote and increasingly offshore, maintenance has become a different kind of challenge. A failed gearbox is no longer simply a mechanical problem. It can result in prolonged downtime, lost electricity generation, expensive crane operations and repairs that are constrained by weather and site access. Preventing failure therefore becomes as important as repairing it.

This is where AI makes its entrance.

Modern wind turbines continuously generate vast quantities of operational data. Sensors monitor temperatures, vibration, rotor speed, wind conditions, oil quality and hundreds of other measurements every few seconds. Machine learning models analyse these streams of information, searching for subtle patterns associated with previous failures. A turbine that appears to be operating normally may already be classified as presenting an elevated risk of failure months before any fault becomes visible to an engineer.

This is often described as predictive maintenance (PdM). But prediction is only part of the story.

The technology does not repair turbines. Engineers continue to replace components, inspect equipment and determine whether maintenance recommendations should be followed. Historical maintenance records, engineering expertise and operational judgement remain fundamental. AI changes how failure becomes visible. It means that instead of waiting for damage to reveal itself mechanically, infrastructure is increasingly interpreted through statistical relationships within data.

Could a wind farm operate without this form of AI?

Absolutely.

Wind farms did so for many years, and many continue to rely upon scheduled inspections and conventional monitoring systems. The turbines would continue generating electricity. Maintenance would continue. Components would still be repaired or replaced.

What changes is how uncertainty is managed.

Instead of responding to failures after they occur, operators increasingly intervene before failure becomes material. Maintenance shifts from being primarily reactive to increasingly anticipatory. The turbine is no longer understood only as a physical machine composed of blades, bearings and gearboxes. It also becomes a stream of probabilities, predictions and future risks that require continual interpretation.

This may appear to be a relatively modest technical innovation. However, it quietly changes the character of renewable infrastructure. Caring for a wind farm increasingly means caring through prediction. Reliability is no longer secured solely through engineering and repair, but through computational systems that continually anticipate futures before they arrive.

Again, the question is not whether renewable energy requires AI. It does not. The more interesting question is what happens when maintaining critical infrastructure increasingly depends upon computational models that make future failures visible in the present.

Does a Wind Farm Need AI to Predict the Wind?

Once again, the short answer is no.

Wind farms have always depended upon weather forecasting. Long before machine learning entered the energy sector, meteorologists used atmospheric observations, numerical weather prediction models and statistical forecasting to estimate wind speeds and weather conditions. Operators planned maintenance around expected weather windows, while grid operators balanced electricity supply against anticipated changes in generation. Renewable energy has never required AI to anticipate the weather.

So why has forecasting become increasingly computational?

The answer lies not in predicting the weather, but in predicting electricity.

A wind farm needs to know far more than whether tomorrow will be windy. It needs to estimate how much electricity each turbine is likely to generate over the coming hours, how confident those predictions are, and how that generation will align with maintenance schedules, electricity markets and wider demand across the grid. As renewable energy has expanded, these questions have become increasingly complex.

Machine learning is now frequently combined with meteorological forecasting to improve these predictions. Weather models are integrated with historical generation data, turbine performance, local topography and operational information to estimate future electricity production. Instead of asking simply How strong will the wind be?, forecasting increasingly asks How much electricity will this wind farm generate, when, and with what degree of uncertainty?

Again, AI is not replacing existing expertise.

Meteorologists continue developing atmospheric models. Engineers continue interpreting turbine performance. Grid operators continue deciding how much uncertainty the electricity system can accommodate. Machine learning extends these practices by identifying relationships within increasingly large and complex datasets. Forecasting becomes a collaborative achievement between atmospheric science, engineering expertise and computational prediction.

Could a wind farm operate without these systems?

Certainly.

Wind farms generated electricity long before machine learning forecasting became commonplace, and electricity systems have always accommodated uncertainty. Less accurate forecasts do not prevent renewable energy from functioning. They simply make the system less efficient. Maintenance becomes harder to schedule. Electricity markets become more volatile. Grid operators require larger margins of safety to compensate for uncertainty.

What changes, then, is not the weather.

It is how uncertainty is governed.

The atmosphere becomes a series of computational futures. Forecasts are no longer simply descriptions of tomorrow’s weather; they become operational knowledge through which decisions are made today. Maintenance is scheduled before the wind arrives. Electricity is traded before it is generated. Grid operators prepare for conditions that do not yet exist.

Following forecasting in this way reveals another quiet inheritance of climate transition. Renewable energy increasingly depends not only upon turbines and transmission lines, but upon computational systems that continually translate uncertain futures into present decisions. The question is not whether we can predict the weather without AI. We can. The more interesting question is what kinds of computation we are prepared to embrace. Few would object to computational systems that reduce forecasting uncertainty, improve the integration of renewable electricity or lower the cost of decarbonisation. Yet these are precisely the kinds of AI that often disappear from public debate. Instead of asking whether AI is simply good or bad, climate transition asks us to distinguish between the different forms of computation we should and should not be building into the infrastructures of everyday life.

Does a Renewable Grid Need AI?

This is where the answer becomes more complicated.

For much of the twentieth century, electricity systems were comparatively predictable. Large coal and gas power stations generated electricity continuously, increasing or decreasing output in response to changes in demand. Electricity flowed in one direction, from power station to consumer, and grid operators balanced the system using established operational procedures, engineering expertise and increasingly sophisticated, though largely deterministic, computational models.

That system no longer exists.

Renewable energy is transforming the grid into a far more dynamic environment. Electricity generation now fluctuates with changing weather conditions. Offshore wind farms, solar arrays, battery storage, interconnectors, electric vehicles and millions of distributed energy resources all interact simultaneously. Electricity no longer simply flows across the network. It is continually generated, stored, exported, imported and redistributed across multiple scales.

Coordinating this system is no longer only an engineering challenge. It is an informational one.

Increasingly, machine learning and optimisation systems bring together forecasts of renewable generation, electricity demand, battery capacity and network conditions to support decisions about balancing the grid. Which batteries should discharge first? Should electricity be imported through an interconnector? Is renewable generation likely to meet evening demand, or should additional generation be brought online? Rather than governing a single wind turbine, these systems increasingly coordinate relationships across an entire electricity network.

Could a renewable electricity system operate without AI?

In principle, yes.

Electricity grids existed long before machine learning became available, and system operators have managed uncertainty for decades through forecasting, operational experience and conventional optimisation techniques. There is nothing inherently impossible about operating a renewable grid without AI.

The more difficult question is whether it remains practical as renewable energy continues to expand.

Every additional wind farm, battery, electric vehicle and distributed generator increases the complexity of balancing the system. Decisions must be made more frequently, across more variables and under greater uncertainty than ever before. Computation does not eliminate that complexity. It becomes one of the primary ways through which complexity is managed.

What changes, then, is not simply the electricity grid.

It is the infrastructure upon which decarbonisation increasingly depends.

This is the paradox that climate debates rarely acknowledge. Data centres have become powerful symbols of digital excess, criticised for their electricity consumption, water use and expanding physical footprint. Yet some of the computation they make possible is becoming embedded within the everyday operation of renewable energy itself. The systems that many campaigners associate with environmental harm are, at least in part, also helping coordinate the infrastructures expected to reduce our dependence on fossil fuels.

This does not mean that every form of AI is environmentally necessary, nor that all computation should be accepted uncritically. Recommendation algorithms, advertising platforms and critical energy infrastructure are not equivalent forms of computation. But recognising this distinction changes the conversation. The question is no longer whether we should reject AI wholesale. It is which forms of computation become necessary, for whom, and under what conditions.

Perhaps this is another Carbon Afterlife.

Fossil fuels left behind pipelines, refineries and atmospheric carbon. Renewable energy may leave different inheritances. Sensor networks, forecasting systems, optimisation models, cloud infrastructure and new forms of computational dependency. These too will require maintenance, governance and care. The transition away from fossil fuels is not merely replacing one energy system with another. It is assembling another kind of infrastructure, one whose afterlives we are only beginning to recognise.

What Kind of Computation Are We Building?

This essay began with what appeared to be a straightforward question.

Can we run a wind farm without AI?

The answer, in almost every case, is yes.

We protected birds before computer vision. We maintained turbines before predictive maintenance. We forecast weather before machine learning. Electricity grids balanced supply and demand long before contemporary AI systems emerged. Renewable energy does not suddenly cease to function without artificial intelligence.

At the same time, each of these examples also reveals something else.

As renewable infrastructures become larger, more distributed and increasingly interconnected, computation is becoming embedded within their everyday operation. AI is not replacing engineers, ecologists or meteorologists. It is becoming another way through which environmental care, maintenance, forecasting and coordination are organised. The question then becomes, what kinds of computation climate transition is quietly assembling.

This matters because debates about AI have become increasingly polarised.

Governments argue that AI is essential if nations are to remain economically competitive. Data centres promise investment, employment and technological leadership. Rejecting them, we are told, risks leaving countries behind in the next industrial revolution.

Environmental critics often begin elsewhere. Data centres consume vast quantities of electricity, water and land. They become another form of extractive infrastructure whose environmental costs appear incompatible with climate commitments.

Both positions contain important truths.

Neither asks what computation is actually for.

The AI that recommends advertisements, generates endless streams of online content or keeps us scrolling through social media is not equivalent to the computational systems that help protect birds, anticipate equipment failure or coordinate renewable electricity across an increasingly complex grid. Treating computation as though it were a single environmental object obscures the different worlds it helps assemble.

Perhaps this is another Carbon Afterlife.

The transition away from fossil fuels is not only replacing one source of energy with another. It is also creating new infrastructures of sensing, prediction and coordination that future generations will inherit alongside turbines, batteries and transmission lines. These computational inheritances will require governance, maintenance and care just as surely as the physical infrastructures they support.

The question, then, is no longer whether we are for or against AI.

Nor is it simply whether we are for or against renewable energy.

It is what we are saying yes to when we say yes to transition.

If renewable energy increasingly depends upon computation, then climate politics can no longer treat digital infrastructure as something separate from environmental transition. Equally, if not all computation serves the same purposes, then we need a richer political conversation about which forms of computation are socially and ecologically necessary, which merely accelerate consumption, and who gets to decide the difference.

Perhaps that is the question we have been avoiding all along.

x

As international attention once again turns towards Iran through the urgency of war, another transformation continues largely unnoticed in the country's northwest. Lake Urmia's disappearance is often narrated as an environmental tragedy. However, I think the lake also reveals something larger about the worlds climate transition is already creating…

Before Lake Urmia became an environmental crisis, it belonged to another geography.

Growing up, I encountered the lake long before I knew its hydrology. It appeared in the landscapes of Assyrian history that shaped my family’s story – villages scattered across northwestern Iran, shifting borders, episodes of displacement, and journeys that stretched towards the Caucasus and beyond. Long before I came across Lake Urmia in satellite imagery or climate reports, it was already part of a landscape marked by empire, migration and survival.

Today, the lake is known differently.

Over the past three decades, Lake Urmia has shrunk dramatically. Once the largest salt lake in Central Asia, it has become an international symbol of ecological collapse. Scientists point to drought, dam construction, intensive irrigation and rising salinity; satellite images trace the lake's retreat with striking clarity.

But that is not the only story the lake tells.

As the water has receded, another landscape has come into view. Restoration projects have multiplied. Satellites and hydrological models now monitor the lake with remarkable precision. Governments, scientists and international organisations have assembled new infrastructures of environmental governance. More recently, speculation has turned towards the lithium believed to lie beneath the exposed salt flats—a mineral increasingly framed as essential to the global transition away from fossil fuels.

That paradox is what drew me back to Lake Urmia. I think of these as carbon afterlives. By carbon afterlives, I do not mean what remains once fossil fuels disappear. I mean the political, ecological and economic worlds assembled through governing carbon persistence. Instead of asking only whether emissions decline, carbon afterlives ask what kinds of infrastructures, responsibilities, territories and futures are assembled as carbon is relocated, contained and governed beyond the point of emission. Lake Urmia offers one way of following those worlds as they emerge.

We often imagine environmental collapse and climate transition as opposites – first a landscape is damaged, then it is repaired or left to be a sacrifice. But the closer I followed the lake, the more that story unravelled. The same drying landscape that demands restoration also becomes newly valuable. Conservation, monitoring and critical mineral speculation do not replace one another; they accumulate. Environmental loss becomes the condition for new forms of scientific attention, political authority and economic value.

Lake Urmia is therefore not only a story about a disappearing ecosystem. It is a story about the kinds of world’s climate transition builds.

That matters because climate transitions are often narrated as morally straightforward. Fossil fuels give way to renewables. Degradation gives way to restoration. Carbon emissions fall, and the future becomes cleaner. Yet transitions unfold within landscapes already shaped by colonial histories, uneven development and global capitalism. From debates over differentiated responsibility in international climate negotiations to conflicts over water allocation within river basins, climate politics has always been about how responsibilities and sacrifices are distributed rather than simply how emissions are reduced.The infrastructures we build to repair the climate inherit those histories. They redistribute value, responsibility and power as much as they reduce emissions.

Following Lake Urmia made me realise that climate change does not only leave damaged landscapes behind. It leaves new worlds in its wake. To understand the politics of transition, we have to follow those worlds as carefully as we follow carbon.

Following the Water

Every environmental crisis asks us to follow something.

In Lake Urmia’s case, the obvious answer is water. The shoreline retreats. Rivers run dry before they reach the basin. Salt replaces wetlands. The temptation is to see the lake as the object of concern—as though the crisis begins and ends where the water once stood.

But water refuses those boundaries.

The lake is only the visible end of a much larger hydrological system stretching across thousands of square kilometres. Rivers descending from the Zagros Mountains, reservoirs, irrigation canals, groundwater aquifers and agricultural land are all entangled in the lake’s fate. Over the past few decades, more than sixty dams have been built across rivers feeding the basin, while expanding irrigation and groundwater extraction have steadily reduced the water reaching the lake. Climate change has intensified these pressures, but it has not acted alone. Lake Urmia did not simply dry because rainfall declined. It dried through decades of political decisions about agriculture, infrastructure and development.

That matters because following the water quickly leads away from the lake.

Water diverted towards agriculture sustains farms and livelihoods across the basin. Restoring the lake therefore means redistributing scarcity as opposed to creating abundance. Every proposal to return water to the lake raises another question, where should that water come from, and who should bear the cost? Environmental restoration becomes inseparable from rural livelihoods, food production and regional inequality.

This is where the language of transition begins to feel deceptively simple.

We often imagine ecological restoration as the opposite of environmental degradation—as though damaged landscapes can be repaired once the right people, technologies or policies are in place. But restoration is never a return to an untouched past. It is a political process of deciding which futures should be made possible and which sacrifices become acceptable. Who continues farming? Which communities adapt? What forms of life are protected, and which are expected to change?

For me, this is where Lake Urmia stopped being a story about water and became a story about power.

The lake I had inherited through family memory was never outside politics. Long before climate change, these landscapes had been shaped by empire, shifting borders, forced migration and uneven development. Climate transition did not arrive on an empty stage. It entered a place where histories of dispossession had already determined who controlled land, water and mobility.

Following the water therefore reveals something larger than a hydrological crisis. It reveals that climate transitions are not simply environmental projects. They reorganise relationships—between states and communities, ecosystems and agriculture, memory and development. They build new worlds from landscapes that were never politically neutral to begin with.

Following the Gaze

As Lake Urmia has changed, so too have the ways people relate to it. The shrinking shoreline has generated an extraordinary apparatus of observation. Satellites trace the lake’s changing extent almost daily. Hydrological models estimate inflows and evaporation. Remote sensing maps salinity, vegetation and dust emissions. Once a regional landscape, Urmia has become an object of continuous global observation.

This visibility matters. Satellite imagery has made the lake's transformation difficult to deny, but monitoring does more than document environmental change. It creates new ways of living with it.

As the lake becomes measurable, it also becomes governable. Restoration targets are established. Funding priorities emerge. International collaborations form around shared datasets and environmental indicators. The lake is no longer only a body of water; it becomes a site through which climate knowledge, expertise and intervention are organised.

This is increasingly characteristic of climate transition.

Responding to environmental change is not simply a matter of building seawalls or restoring wetlands. It also involves building new institutions, media and forms of collective attention through which people learn to inhabit changing environments.

Tuvalu offers a striking example. As rising sea levels threaten the country’s future, the government has not only invested in adaptation but also launched Tuvalu TV and digital nation initiatives that seek to preserve culture, language and political identity in the face of climate change. Climate adaptation here is not only about resisting loss. It is about creating new ways of remaining present in a changing world.

Lake Urmia represents a different kind of transition, but one animated by a similar dynamic. Environmental monitoring does not simply record the lake’s decline. It creates new relationships between scientists, governments, international organisations and local communities. It reshapes what counts as environmental success and who participates in producing that knowledge.

Seen in this way, climate transitions are not only technological or ecological projects. They are social and political processes through which new infrastructures of knowledge, care and governance emerge alongside changing environments.

Of course, every way of seeing also brings its own limits.

Datasets privilege what can be measured. Models simplify complex social worlds. Indicators rarely capture histories of displacement, cultural attachment or everyday experience. This is not a criticism of science—without scientific observation many environmental crises would remain invisible—but a reminder that every infrastructure of knowledge highlights some relationships while leaving others in the background.

Following the gaze therefore reveals that climate transition is not only about responding to environmental change. It is also about assembling new ways of knowing, governing and inhabiting the worlds that climate change is already transforming.

Following the Salt

If you were standing on the shores of Lake Urmia today, it is the salt that catches your attention.

The water has retreated, leaving behind a vast white plain that stretches towards the horizon. Once submerged beneath one of the largest salt lakes in Central Asia, the exposed lakebed has become the defining image of Urmia’s ecological decline. It is easy to read this landscape as absence—as evidence of everything the lake has lost.

But salt tells another story.

Within the hypersaline brines and sediments beneath the lake lie significant concentrations of lithium, a mineral that has become central to the global energy transition. Whether those reserves will ever be extracted remains uncertain. Commercial viability, environmental constraints and political conditions all remain open questions. Yet the possibility has already begun to reshape how the lake is imagined.

The irony is difficult to ignore.

The same environmental crisis that made Lake Urmia a symbol of ecological collapse also made it newly valuable. As demand grows for the materials, carbon sinks and ecosystems needed to support decarbonisation, landscapes once understood primarily through environmental loss are increasingly reimagined as strategic assets. From lithium-rich salt lakes in the Andes to peatlands valued for carbon storage and mangrove forests enrolled in blue carbon markets, climate transition is not simply protecting nature. It is producing new ways of valuing it.

Lake Urmia increasingly belongs to that geography.

This is not simply a story about lithium. It is a story about how environmental value is remade.

Across the world, damaged ecosystems are being enrolled in new economies of climate transition. Forests become carbon sinks. Wetlands become natural infrastructure. Mangroves become blue carbon. Landscapes are valued not only for the lives they sustain, but for the carbon they store, the minerals they contain or the ecosystem services they provide. Conservation and extraction, once imagined as opposites, increasingly coexist within the same political and economic landscape.

That coexistence should give us pause.

The green transition is often narrated as an escape from the extractive logics of the fossil fuel economy. Yet renewable technologies depend on immense quantities of land, water and critical minerals. Decarbonisation does not end extraction; it reorganises it. Nor does ecological restoration exist outside these dynamics. Restoration projects, carbon markets and biodiversity finance all seek to repair environmental damage while simultaneously producing new forms of economic value. Sometimes those goals align. Sometimes they compete. They are never entirely separate.

This is where Lake Urmia begins to challenge the stories we tell about climate action.

The question is no longer how to restore the lake, but what kind of future restoration is building. Who benefits from the new values attached to damaged landscapes? Who bears the costs of conservation, extraction or changing patterns of resource use? Can climate transition avoid reproducing the unequal geographies of wealth and power that have long shaped the global economy, or does it risk giving those inequalities a greener language?

The exposed salt of Lake Urmia is therefore more than the residue of disappearing water. It is the surface upon which competing futures are written. Ecological recovery, scientific intervention, economic speculation and historical memory all converge here, layered one upon another. The lake has not ceased to matter because it is disappearing. It matters because disappearance has made it newly visible, newly valuable and newly contested.

The Lake Remembers

When I began following Lake Urmia, I thought I was writing about environmental collapse.

Instead, I found myself tracing the lives that gather around a disappearing landscape. Water led to agriculture and governance. Satellites led to new ways of seeing. Salt led to lithium, carbon and the political economies of climate transition. The lake kept refusing to remain a single object. It became many things at once – an ecosystem, a homeland, a dataset, a restoration project, a speculative frontier.

Perhaps that is what climate transition looks like.

We often imagine transition as a destination—a world after fossil fuels, where cleaner technologies replace dirtier ones and environmental damage is gradually repaired. But transitions do not simply reduce emissions. They reorganise relationships. They redistribute value. They decide which landscapes should be restored, which should be protected, which should be mined and, ultimately, whose futures those decisions serve.

None of this is an argument against climate action. The climate crisis demands rapid decarbonisation and ecological restoration. But if transition is to be just, we cannot measure success only in tonnes of carbon avoided or batteries produced. We must also ask what kinds of worlds those achievements create, and for whom.

That is why I keep returning to Lake Urmia.

For my family, the lake has never been simply a body of water. It belongs to a longer history of Assyrian life, displacement and survival. Long before satellites mapped its shoreline or investors imagined lithium beneath its salt, it was already a landscape carrying memories of empire, migration and loss. Climate transition did not arrive in an empty place. It entered a landscape already shaped by history.

If one were to stand on the shore today, it would be tempting to see only what has disappeared.

But look a little longer.

The salt is not empty. It is crowded—with restoration plans, satellite images, scientific models, mineral speculation, international funding, local struggles and inherited memories. These are the carbon afterlives of Lake Urmia, not what remains once the water is gone, but what emerges because it is.

The lake remembers.

The question is whether we will remember with it.

x

A few years ago, I had the privilege of helping Aras Amiri bring together an evening called Rira: The Call of Iran’s Environmental Prisoners at Edinburgh University.

It was June 2023, and we gathered scientists, conservationists, artists and campaigners to speak about something that should never have required campaigning at all, the imprisonment of people whose life’s work was protecting nature.

We spoke about the Asiatic cheetah.

About the deserts, forests and mountains of Iran.

About the extraordinary biodiversity of a country too often reduced to headlines about politics.

But most of all, we spoke about the people who had dedicated their lives to protecting it.

Houman Jowkar. Sepideh Kashani. Niloufar Bayani. Taher Ghadirian. Sam Rajabi. Amirhossein Khaleghi. Morad Tahbaz. So many names that deserved to be known for science rather than suffering.

That evening was filled with hope.

There were photographs of wildlife taken before arrests interrupted lives. There was music. There were memories. There was an unwavering belief that telling these stories mattered—that if enough people knew their names, perhaps they would not be forgotten.

Not long afterwards, many of them were finally released.

It felt, if only briefly, as though justice had managed to find a way through.

This week that hope has been shattered.

Houman Jowkar and Sepideh Kashani have reportedly been rearrested, once again taken from their home by Iranian security forces. Once again, their families are left waiting. Once again, those who know them are asking the same impossible questions.

How do you imprison people whose greatest act was to protect one of the world’s rarest cats?

How do you treat conservation as a crime?

How many times must someone lose years of their life for defending the natural world?

I keep thinking back to that evening in Edinburgh. To the scientists who travelled from across the world because they knew these prisoners not as symbols, but as colleagues and friends.

To Aras, who understood imprisonment not as an abstract injustice but as lived experience.

To the photographs of the Asiatic cheetah—images made possible only because someone cared enough to spend years quietly watching, documenting and protecting a species on the edge of extinction.

Those images feel different now.

They remind us that conservation is ultimately an act of hope. You protect something because you believe there will be a future worth protecting.

The tragedy is that the people who embody that hope continue to be treated as threats.

There are moments when words feel painfully insufficient.

This is one of them.

But silence would be worse.

So today I am saying their names again.

Houman Jowkar.

Sepideh Kashani.

And I am remembering everyone who stood beside them, everyone who has refused to let their work be erased, and everyone who believes that caring for wildlife should never lead to a prison cell.

The work they devoted their lives to is bigger than any one arrest.

The least we can do is continue to bear witness.

Because the call of Rira was never only about Iran’s endangered wildlife.

It was—and still is—a call not to look away.

x

More reading:

https://www.theguardian.com/world/2026/jul/13/nazanin-zaghari-ratcliffe-condemns-iran-rearrest-of-wildlife-activists-houman-jokar-sepideh-kashani

https://commapress.co.uk/blog/for-peace

Last week I suggested that Carbon Afterlives would continue at a slower pace while the research developed. Then I started following a cable. One question led to another, and before long there was another essay!

When futures become cables

For many years Orkney has occupied a distinctive place within Britain’s climate imagination. It is regularly presented as a place where another kind of energy future is already emerging, community-owned wind turbines, locally generated electricity, experiments in democratic ownership and an island economy attempting to live beyond fossil fuels. Unlike many climate futures that remain trapped within policy documents or speculative scenarios, Orkney often appears as evidence that another way of organising energy is not only imaginable but already underway.

I have always found these projects compelling. They represent a different political imagination, one in which energy is not only consumed but collectively produced and governed. The future they propose is not merely technological. It is social. Renewable energy becomes inseparable from ideas of community, participation and local autonomy.

Looking through a Carbon Afterlives lens, however, another story gradually begins to emerge.

Orkney now produces considerably more renewable electricity than its local grid can absorb. On particularly windy days turbines are routinely constrained because there is simply nowhere for the electricity to go. The problem is no longer one of generating renewable energy. It is one of moving it. The climate transition, in other words, has encountered the material limits of the infrastructures inherited from an earlier energy system.

The more I thought about this, the more the conversation seemed to shift. It was no longer simply about local generation or democratic ownership. Increasingly it revolved around energy security, national resilience and the need to strengthen Britain’s electricity network in response to a rapidly changing energy landscape. The future being imagined had become larger than the island itself. Wind generated in Orkney would have to travel hundreds of miles to homes and cities elsewhere. Renewable electricity would need to be shared across regions and across national borders.

That future has acquired an extraordinarily material form.

Beneath the North Sea, new subsea electricity interconnectors are beginning to redraw the geography of Britain’s energy system. The Greenlink cable, linking Ireland and England, is one example among a growing network of high-voltage connections that will move renewable electricity between places of abundance and places of demand. Similar energy “superhighway” cables stretching between Scotland and England promise to carry several gigawatts of offshore wind southwards over the coming decade.

Seen from the perspective of climate policy, these projects appear almost self-evident. Renewable electricity must travel. Energy systems must become more flexible. Interconnection improves resilience, lowers electricity costs and strengthens energy security. The future arrives beneath the sea.

Still, if you trace the subsea cable, rather than the electricity, it tells a different transition story.

A cable crossing the seabed is not simply laid across the ocean floor. Before construction begins, years of geological surveys, environmental assessments, fisheries consultations and marine licensing take place. Specialised survey vessels map sediments, archaeological remains and marine habitats. Jet trenching machines excavate the seabed before the cable is buried approximately one and a half metres below the surface to protect it from anchors and fishing gear. At landfall, horizontal directional drilling carries the cable beneath beaches and cliffs before converter stations connect it to the terrestrial electricity network. Construction takes years. Marine sediments are redistributed. Benthic habitats are disturbed. Entire logistical systems are assembled around what ultimately appears on maps as a remarkably simple line.

Reading the technical documents, I found myself thinking back to the final essay in the first Carbon Afterlives series. There I suggested that climate futures do not remain in the future. They become embedded within infrastructures, institutions and everyday obligations long before the futures they promise have fully arrived. Looking at these cables, it seemed to me that this was precisely what I was witnessing. The imagined future of renewable energy had acquired weight. It had become planning inquiries, converter stations, seabed surveys, construction vessels and buried infrastructure. The future was no longer waiting to arrive. It was already being installed.

Following the material

Then I found myself asking a much simpler question.

What, exactly, was this cable made from?

Following the answer proved unexpectedly difficult because the official narrative surrounding interconnectors is overwhelmingly concerned with electricity. Wind becomes renewable power. Renewable power becomes energy security. Electricity moves silently beneath the sea.

The cable, however, has another history.

Increasingly, the copper used in these high-voltage systems does not originate from newly mined ore but from dismantled electrical infrastructures collected across Europe. Yesterday’s substations, transmission lines, transformers and industrial equipment become tomorrow’s renewable grid. This recirculation reflects a much broader shift within European industrial policy. Instead of treating obsolete infrastructure as waste, the European Union increasingly understands it as an urban mine, a strategic reserve of critical raw materials whose value should remain in circulation through successive cycles of recovery and reuse.

The Greenlink cable forms part of this wider industrial ambition. Its manufacturer, Prysmian, has partnered with the Catalan recycler La Farga to produce high-voltage cable using 100% recycled copper. Scrap collected from decommissioned electrical systems across Europe is transported to La Farga’s metallurgical facilities where it is melted at temperatures exceeding 1,000℃, refined and continuously cast into new wire rods before travelling onwards to specialist cable manufacturing plants elsewhere in Europe. Only then does it become the conductor eventually buried beneath the North Sea.

What interested me was not that the copper had been recycled. It was what happened while it was being recycled.

Recovering copper does not simply produce a purified metal. It also produces another material, metallurgical slag. Once regarded primarily as an industrial residue, slag is increasingly being reclassified as a secondary raw material. Working with researchers at the Polytechnic University of Catalonia, La Farga has explored how this by-product can substitute for virgin iron ore and silica sand used in cement manufacture.

One industrial process, in other words, becomes the beginning of another.

Following the slag

Following the slag leads into a very different industrial landscape.

Cement manufacture is responsible for around 8% of global CO₂ emissions and remains one of the most difficult sectors to decarbonise because much of its carbon dioxide is released through the chemistry of clinker production. For this reason, companies such as Holcim and Cemex are investing heavily in carbon capture technologies designed to intercept carbon dioxide before it reaches the atmosphere.

Here, the trail branches again.

Captured carbon dioxide does not always travel directly to long-term geological storage beneath the seabed. Increasingly, some of it enters another industrial system altogether. In the greenhouse agriculture of Almería in southern Spain, purified carbon dioxide is piped into enclosed growing environments where elevated concentrations of CO₂ stimulate photosynthesis, accelerating the growth of tomatoes, peppers and cucumbers during the winter months. The vegetables are harvested, packed and transported northwards through European distribution networks before appearing on supermarket shelves across Britain.

Some days after tracing this chain, I found myself standing in front of a packet of Spanish tomatoes.

It was difficult not to think about the extraordinary journey contained within such an ordinary object. A transition that had begun with community-owned wind turbines in Orkney and a cable buried beneath the North Sea had unexpectedly led me to metallurgical furnaces in Catalonia, cement chemistry, greenhouse horticulture in southern Spain and finally a supermarket refrigerator thousands of kilometres away.

The carbon captured from an industrial process would shortly become part of someone else’s metabolism. Through photosynthesis it would be incorporated into the tomato, digested, circulated through the bloodstream and eventually metabolised back into carbon dioxide before returning to the atmosphere through respiration. Some of it would ultimately dissolve into the oceans. Some would enter other biological systems. Some might once again become part of another industrial process.

The transition, I realised, was not producing clean endings.

It was producing new circulations.

Stories that materials exceed

Standing in the supermarket, it occurred to me that this was not simply a story about hidden supply chains. It was also about the limits of the stories we tell ourselves about climate transition.

Every organisation I had encountered along the way possessed its own coherent account of what it was doing. In Orkney, the story was one of community ownership, local benefit and renewable abundance. For transmission operators, it became one of energy security and national resilience. For cable manufacturers, recycled copper represented the circular economy in practice. For metallurgists, industrial residues became secondary raw materials. For cement producers, carbon capture offered a route towards decarbonising one of the world’s most difficult industries. For greenhouse growers in Almería, carbon dioxide became another agricultural input. None of these narratives was wrong. At the same time, none was capable of containing the remarkable material world that emerged when they were placed alongside one another.

Looking at those tomatoes on the shelves, it seemed to me that following the cable had revealed something else as well. Climate transition does not unfold through a single shared narrative. It unfolds through overlapping socio-technical imaginaries, each with its own politics, institutions, ambitions and measures of success. Community energy, energy security, circular manufacturing, industrial decarbonisation and greenhouse productivity each describe only one part of the transition. The materials themselves, however, refuse these boundaries. Copper, slag, carbon dioxide and tomatoes move effortlessly between them, assembling relationships that exceed the ambitions, politics and temporal horizons of any individual actor.

Delay

Perhaps this also changes how we think about delay.

Climate politics often treats delay as something to be overcome, delayed emissions reductions, delayed implementation, delayed action. Yet following the cable suggested another temporal logic altogether. Here, delay is not simply political. It becomes material.

Recycling the copper does not eliminate the need for copper. It extends the life of material that has already been extracted, allowing yesterday’s electrical infrastructure to remain in circulation rather than returning immediately to waste or demanding new mining. The impurities removed during recycling do not disappear either. They become metallurgical slag, reclassified as a secondary raw material before entering cement manufacture instead of landfill. Carbon captured from industrial processes does not simply vanish into geological storage. In some cases it is redirected into greenhouse horticulture where, for a time, it becomes part of a living plant before entering our own bodies through food, returning to the atmosphere through respiration and eventually dissolving into the sea or circulating elsewhere.

At every stage, matter is held for a while longer. It is given another function, another destination, another temporary life before moving on again.

Sara Ann Wylie describes such movements as existing in "relations of becoming", where materials pass continually between bodies, infrastructures and environments, "each shaping the conditions of those yet to come." I have found myself returning repeatedly to that final phrase. Following the cable suggests that climate transition is assembled through precisely these unfinished relations. Materials are not simply extracted, recycled or stored. They continue into other processes, other landscapes and other bodies, carrying with them responsibilities and possibilities that exceed the purposes for which they were first mobilised.

Perhaps this is another way of understanding Carbon Afterlives. It is not simply about what happens to carbon once it has been emitted, captured or stored. It is about what becomes visible when we follow materials beyond the point where one story ends and another begins. The future, it turns out, is not waiting patiently beyond the transition. It is already circulating through cables buried beneath the North Sea, recycled copper, cement plants, greenhouse tomatoes, our own bodies and, eventually, back into the atmosphere and the sea—each shaping the conditions of those yet to come.

x

Inheriting Transition

For the past two decades, climate politics has generated a remarkable proliferation of futures. From Transition Towns, Transition Design and Just Transition to regenerative cultures, Solarpunk and more-than-human imaginaries, scholars, activists and practitioners have developed diverse ways of imagining life beyond fossil dependence.

Elsewhere, pluriversal and decolonial perspectives have challenged the universalising assumptions of Western development, while Indigenous resurgence, climate fiction, speculative design and participatory foresight have expanded the horizons of political imagination. Doughnut economics, post-growth, post-carbon and Green New Deal imaginaries have each offered competing visions of how economies and societies might be reorganised under conditions of ecological constraint.

These projects differ profoundly in their politics, methods and assumptions, yet they share a common commitment to futurity. Each begins from the premise that climate politics requires the capacity to imagine otherwise.

The importance of these imaginaries is difficult to overstate. At a moment when climate discourse has often appeared trapped between environmental catastrophe and technological optimism, these projects have challenged fatalism and reopened political imagination. They have expanded the horizons of political possibility, creating space for experimentation, rehearsal and the cultivation of different relationships between people, technologies and ecologies.

Earlier projects of my own, including Sunlight Doesn’t Need a Pipeline, emerged from this tradition. I was interested in how everyday climate realities might be made visible differently, and how imagination could disrupt dominant narratives of transition and create space for alternative ways of living together. I continue to believe this work matters.

At the same time, alongside these projects, another line of inquiry was quietly taking shape.

Over the past several years, while researching pollution, waste, chemical exposure, climate governance, carbon management and energy transitions, I found myself returning repeatedly to substances, infrastructures and obligations that do not simply disappear. Creating artistic projects and reading across work on toxicity, uncertainty and life in damaged worlds gradually shifted my attention. Instead of asking only what kinds of futures climate politics makes possible, I became increasingly interested in the inheritances it is already assembling.

What, exactly, is being secured through contemporary climate governance? What forms of life are being maintained, and at what cost? What responsibilities are being created in the name of transition? And what happens when the future is no longer imagined primarily as transformation, but as the continuation of an existing order under conditions of escalating instability?

I do not think these questions replace the politics of imagination. To me, they suggest that climate politics may have entered another phase. Alongside imagining different worlds, we are increasingly confronted with the task of inhabiting futures that are already becoming material. The challenge is no longer only to imagine otherwise, but to recognise how imagined futures settle into infrastructures, institutions and everyday obligations long before the futures they promise have fully arrived.

Endurance

A few months ago, I found myself reading technical reports about the future stewardship of carbon stores beneath the North Sea. Among them was Endurance, the geological reservoir intended to store millions of tonnes of carbon dioxide beneath the seabed.

The reports are, for the most part, entirely unremarkable. They contain diagrams, monitoring protocols, regulatory frameworks and assessments of long-term risk. Their language is procedural rather than visionary. They concern questions of measurement, responsibility and oversight. Who will monitor the site? For how long? Which institutions will remain accountable? What happens if leakage occurs decades after injection has ceased?

The more time I spent with these documents, the stranger they became.

Carbon storage is usually presented as a climate technology, one mechanism among many for reducing atmospheric emissions on the pathway towards net zero. Viewed from another angle, however, a carbon store is also an inheritance.

Once carbon enters a geological formation, it does not simply disappear. Atmospheric concentrations may be reduced, but new obligations emerge in their place. Monitoring systems must be maintained. Regulatory frameworks must endure. Knowledge must be preserved and transmitted. Responsibilities must be distributed across institutions that may themselves change, fragment or disappear over time.

The longer I sat with these documents, the less they appeared to describe a technology and the more they resembled instructions for stewardship. They were not simply concerned with storing carbon. They were concerned with maintaining relationships across time, between infrastructures and institutions, between present decisions and future responsibilities, between those who benefit from climate interventions today and those who will inherit the obligation to care for them tomorrow.

Seen from the perspective of the storage site, the climate transition begins to look rather different. It is no longer simply a pathway towards a low-carbon future. It is also the creation of socio-material relationships that must themselves be sustained.

The carbon remains. The infrastructure remains. The obligation remains.

Following Carbon

Around the same time, I found myself writing about Aberdeen’s hydrogen buses, a project frequently presented through the familiar language of innovation, demonstration and future potential.

Hydrogen promised a pathway beyond fossil fuels. It offered a narrative of regional renewal and a means of carrying the expertise, infrastructures and labour force of the North Sea oil and gas sector into a low-carbon future. Across policy documents, investment strategies and regional development plans, hydrogen appeared as a bridge between worlds; a technology capable of preserving industrial continuity while simultaneously promising transformation.

What increasingly interested me, however, was not only the future being imagined, but the material world already gathering around that imagination.

Refuelling stations appeared across the city. New technical expertise was cultivated. Investment flowed towards infrastructures whose significance depended upon futures that had not yet arrived. Relationships formed between local authorities, industry partners and regional development agencies. Maintenance regimes were established. Regulatory frameworks expanded. Entire ecologies of expectation began to assemble around a future that remained, in many respects, uncertain.

Whatever role hydrogen ultimately comes to play in the energy transition, it had already begun reshaping the present.

Once I noticed this, I began seeing the same pattern elsewhere.

Carbon removal projects were reorganising landscapes around future promises of sequestration. Overshoot pathways embedded within integrated assessment models were shaping present-day policy through assumptions about future removals. Insurance companies were withdrawing from flood-prone regions long before projected sea-level rise reached its anticipated thresholds. Supply chains for lithium, cobalt and rare earth minerals were expanding in anticipation of electrified futures, while adaptation deficits were already determining who would be protected from climate impacts and whose losses would quietly become normalised.

These systems were not speculative.

They already existed.

They had budgets, investors, planning frameworks, legal agreements and physical infrastructures. They occupied landscapes. They reshaped institutions. They redistributed risk and responsibility. Long before their promised futures had arrived, they had already begun reorganising the conditions of everyday life.

Gradually, I realised that I was no longer following carbon alone.

I was following the worlds assembling around it.

Deferred World-Making

Climate transitions are usually narrated through the language of horizons, pathways and destinations. They encourage us to ask what kinds of futures might become possible and what transformations remain ahead.

Following carbon gradually led me towards a different question.

What worlds are already being made?

Not imagined.

Not anticipated.

Made.

What infrastructures are already taking shape? What obligations have already been created? What responsibilities have already been inherited? Which relationships between people, technologies, institutions and ecologies have already become materially embedded through decisions justified in the name of future climate action?

These questions emerged slowly through the Carbon Afterlives series. Looking back, I realise that what changed was not the object of my research but the direction of my attention. Following carbon increasingly meant following the institutions, infrastructures and practices of stewardship that accumulate around substances, emissions and wastes that refuse to disappear.

I have come to think of this process as Deferred World-Making.

The concept emerged from a simple observation. Again and again, climate solutions justified in relation to the future were already acquiring material force in the present.

Future carbon removals legitimise emissions today.

Future geological storage underpins contemporary extraction strategies.

Future adaptation shapes planning decisions before protective infrastructures exist.

Future technologies attract investment, regulation and institutional capacity long before they become technically or economically viable.

In each case, the future enters the present not merely as an idea or a prediction but as a material force.

It becomes embedded within investment decisions, planning systems, legal frameworks, technical expertise and physical infrastructures. It shapes what is funded and what is deferred; what becomes politically imaginable and what quietly disappears from view.

Seen in this way, the future is not waiting patiently at the end of the transition.

It is already being sedimented within it.

This is not to suggest that climate futures are somehow unreal or that imagination has failed. Quite the opposite. It is to recognise that imagination has consequences. Once futures become embedded within infrastructures, institutions and systems of governance, they cease to be speculative alone. They begin organising social and material life in the present, creating obligations that will themselves become part of the worlds future generations inherit.

Climate transitions, in other words, do not simply produce futures.

They also produce inheritances.

Inheriting Transition

Viewed from this perspective, many of the contradictions that animate contemporary climate politics begin to look rather different.

Zambia’s relatively low-carbon energy system cannot be understood apart from histories of colonial extraction, uneven industrialisation and constrained developmental possibility. Equally, debates around nuclear power cannot be separated from the infrastructures, geopolitical dependencies and energy insecurities that earlier energy transitions have already left behind. Neither example offers a simple lesson about the “right” transition. They remind us that climate interventions never unfold in isolation. They become woven into worlds that continue to gather around them, generating new dependencies, responsibilities and possibilities that exceed the futures through which they were first imagined.

Looking back, I think this is what Carbon Afterlives has been trying to follow all along.

Not carbon alone, but the socio-material worlds that accumulate around climate interventions once they leave the page of the scenario, the model or the policy document and enter landscapes, institutions and everyday life.

Climate futures do not remain in the future.

Some become investment decisions.

Others become planning frameworks, geological repositories, hydrogen refuelling stations, insurance markets or carbon budgets.

They acquire material form.

They begin organising relationships between people, places, technologies and ecologies long before the futures they describe have fully arrived.

Perhaps this is where climate imagination has another contribution to make.

For the past two decades, much of its work has been to expand political possibility; to help us imagine worlds beyond fossil capitalism, colonialism and ecological destruction. That work remains indispensable.

But another task is now emerging.

As imagined futures become infrastructures, institutions and inherited responsibilities, imagination must also help us recognise the worlds already taking shape around us. It must help us notice how yesterday’s scenarios become today’s planning systems, how technological promises become maintenance regimes, how visions of transition become obligations that others will inherit.

This is not a call to abandon futurity.

It is an invitation to think differently about its consequences.

Every climate intervention assembles a world.

Some worlds will endure far longer than the futures through which they were first justified.

The question, then, is no longer only what future we wish to create.

It is also what inheritances we are already putting into motion.

A Final Note

This essay marks the final instalment of the Carbon Afterlives series.

What began as an attempt to follow carbon beyond the point of emission gradually became something else, an exploration of the material lives of climate futures, of the ways they settle into infrastructures, institutions, landscapes and everyday obligations long before they arrive as the futures they once promised.

The series will continue, although at a slower pace, as the research develops.

Looking back across these essays, I realise that I was never only following carbon.

I was following the worlds that gather around it.

That realisation has changed how I think about climate imagination.

Its value lies not only in opening possibilities for worlds that do not yet exist. It also lies in helping us remain attentive to the worlds our imagined futures are already bringing into being.

Perhaps this is what it means to inherit a transition.

Not simply to receive the consequences of earlier climate decisions, but to recognise that every intervention—however well intentioned—creates new relationships, new obligations and new worlds that others will one day have to inhabit.

The work of climate politics, then, is not only to imagine better futures.

It is to become careful ancestors of the futures that are already becoming someone else’s present.

x

My thanks to all my friends engaged in artistic, climate, environmental and material futures research. Your attention to the environmental and social costs of transition, the politics of technological promises, and the worlds that emerge in the aftermath of those promises has been an important influence on my thinking. Thank you.

The entrance to Finland’s Onkalo repository sits quietly within an unassuming forest on Olkiluoto Island. Beneath the granite bedrock, engineers have spent decades constructing one of the most ambitious infrastructure projects in human history, a repository designed to contain spent nuclear fuel for up to one hundred thousand years.

The timespan is difficult to comprehend. Agriculture emerged roughly twelve thousand years ago. Written language appeared around five thousand years ago. Entire civilisations have risen, flourished and disappeared within a fraction of the period that Onkalo is intended to remain secure. The challenge confronting its designers is not simply technical it is temporal. How do you communicate danger to people who may not share your language? How do you maintain responsibility across political systems that do not yet exist? What institutions can survive long enough to care for materials whose lifespan exceeds recorded history?

Nuclear waste forces a confrontation with a problem that modern societies often prefer to avoid. Some things do not go away.

For decades, climate politics has largely been organised around the opposite assumption. Carbon emissions are reduced. Carbon is captured. Carbon is offset. Carbon is removed. The language suggests movement towards absence, as though the goal is to make carbon disappear from political concern altogether. The more closely one looks at the infrastructures emerging around net zero however, the more familiar they begin to appear. Pipelines. Storage reservoirs. Monitoring systems. Verification frameworks. Long-term stewardship agreements. Geological repositories. Increasingly, climate governance resembles waste management. Carbon and waste are different in many ways, but where they do converge is that both create obligations that persist long after the moment of production has passed.

The First Custodians

The history of nuclear waste is also a history of unequal responsibility. Across North America, uranium extracted from Indigenous territories helped fuel the nuclear age. Throughout the twentieth century, thousands of uranium mines operated across Diné lands in the American Southwest. The uranium supplied military programmes, nuclear weapons production and civilian energy systems that powered distant cities and industries. The benefits travelled whilst the consequences largely remained.

Abandoned mines contaminated water sources. Radioactive dust circulated through homes and communities. Workers and families experienced elevated rates of illness associated with prolonged exposure. Long after extraction ceased, many sites continued to require monitoring, remediation and care. The story is often told as a story of contamination, but it is also a story about stewardship. Communities became responsible for living alongside the afterlives of an industry whose benefits were largely realised elsewhere.

This pattern is hardly unique to uranium. Throughout environmental history, unwanted materials have rarely disappeared. More often they have been relocated. Burdens move while responsibilities remain. The communities asked to live with contamination, extraction or storage frequently have the least influence over the systems that produced them.

Carbon afterlives increasingly generate similar geographies. Carbon capture projects require host communities. Geological storage reservoirs require monitoring. Offset schemes enlist forests, landscapes and territories into systems of climate management extending across generations. Carbon may be removed from one place, but its afterlives settle elsewhere. The question of where carbon goes is therefore inseparable from the question of who becomes responsible for it.

Life in the Ruins

A different perspective on persistence emerges six thousand miles away in northern Ukraine. Nearly four decades after the Chernobyl disaster, forests have reclaimed abandoned villages. Wolves move through former settlements. Lynx, elk and wild horses now inhabit landscapes once defined by intensive human activity. From above, parts of the exclusion zone appear almost indistinguishable from wilderness. The image is often presented as a paradox. How can life flourish in a place associated with catastrophe?

Environmental humanities scholars have spent years grappling with precisely this question. In Arts of Living on a Damaged Planet, Anna Tsing and her collaborators argue that damaged landscapes are never simply dead landscapes. New forms of life continue to emerge amidst contamination, disruption and ecological change. The Anthropocene is populated not only by infrastructures and pollutants, but by ghosts, monsters and unexpected survivals. Chernobyl illustrates this with unusual clarity.

The exclusion zone is frequently described as abandoned. Nonetheless, abandonment is not quite the right word. The area remains monitored, mapped, regulated and studied. Radiation persists. Scientific institutions persist. Infrastructures of management persist. Ecological life persists. The disaster did not end history but generated new histories.

This is one of the most unsettling lessons offered by damaged environments. Waste rarely represents a final chapter. Instead, it produces new ecological, political and social realities that must be inhabited long after the original event has passed. Persistence becomes a condition of life.

Three Hundred Years

Climate change increasingly poses similar questions. Recently, climate researchers exploring long-term sea-level rise scenarios arrived at a striking conclusion. Even if humanity succeeds in reaching net zero within the coming decades, limiting long-term sea-level rise may require carbon dioxide to be actively removed from the atmosphere for centuries afterwards. In some scenarios, carbon dioxide removal will need to happen for three hundred years.

The timescale feels extraordinary. Much contemporary climate policy is organised around targets for 2030, 2040 or 2050. Governments struggle to maintain commitments across electoral cycles lasting only a few years. However, the emerging logic of carbon removal suggests a future in which societies may be required to sustain systems of carbon management across centuries. The implications are profound.

A three-hundred-year carbon removal programme is not simply a climate intervention. It is an institutional project. Pipelines must be maintained. Reservoirs must be monitored. Carbon accounting systems must remain credible. Storage sites must be governed. Financial mechanisms must endure. Technical expertise must be transferred across generations. The challenge begins to resemble the management of nuclear waste more than the deployment of a new technology.

The question cannot only be when will the carbon can be removed. It becomes, who will spend centuries looking after the infrastructures designed to remove it.

The Maintenance Society

The future imagined by climate politics is often one of transition. Fossil fuels will be replaced, emissions reduced and cleaner systems brought into being. But many of the technologies now proposed as climate solutions suggest a different horizon. Maintenance may become one of the defining political questions of the coming century.

Nuclear energy illustrates this most clearly. For decades, radioactive waste repositories stood as symbols of an impossible inheritance. Facilities such as Finland’s Onkalo repository were designed around a simple but unsettling fact, some materials remain politically active long after the societies that produced them have disappeared. As decarbonisation pressures intensify, nuclear power has been re-evaluated by many as an indispensable climate technology. In France, low-carbon electricity has long depended upon precisely the infrastructures of stewardship that earlier generations sought to avoid. In the United Kingdom, debates that once focused on how to manage nuclear decline have increasingly shifted towards how to finance and govern a new generation of reactors. The problem of stewardship has not disappeared; it has been reframed as a condition of decarbonisation.

A similar logic is beginning to emerge around carbon. Proposed storage projects beneath the North Sea, carbon removal facilities in Iceland and offset forests distributed across multiple continents all depend upon forms of monitoring, verification and care that extend far beyond the moment of intervention. Climate solutions increasingly generate their own afterlives. The challenge, then, may not simply be transition. It may be stewardship. Future generations will inherit not only a changing climate, but the repositories, infrastructures and institutions built to manage it.

The climate crisis is frequently imagined as a problem of transformation. Fossil fuels will be replaced by renewables. Carbon emissions will decline. Cleaner systems will emerge. This vision is not wrong, but it is incomplete. Alongside transition, a parallel landscape of stewardship is beginning to emerge. Carbon repositories, offset forests, monitoring systems, adaptation infrastructures and geological storage facilities all point towards a future increasingly organised around the management of persistence.

The task is not simply changing systems. It is caring for what remains. This perspective reveals something valuable. Carbon afterlives are not merely residues of previous activities. They are infrastructures of obligation. They create responsibilities that extend beyond the lifetimes of those who build them. Like nuclear repositories, they require institutions capable of surviving their creators.

In this sense, climate politics becomes inseparable from questions of inheritance. Future generations won’t just inherit climate change, they will also inherit the responsibility for managing our attempts to solve it.

Living with What Remains

The promise of modern waste systems has often been remarkably simple. Things leave. The rubbish truck arrives. The landfill sits beyond the city. The smokestack disperses emissions into the atmosphere. Waste disappears from view and with it, apparently, from concern.

Climate change has steadily undermined that promise. Carbon accumulated in the atmosphere long after it vanished from sight. The consequences returned through altered climates, rising seas and shifting ecological systems. What appeared absent remained active.

The same tension runs through nuclear repositories, contaminated landscapes and abandoned extraction sites. Places such as Onkalo, Chernobyl and the uranium mines scattered across Indigenous lands remind us that waste is rarely a problem that societies solve once and for all. More often, it becomes a relationship that must be organised, inherited and maintained. Materials persist. Institutions persist. Responsibilities persist.

The question, then, is not only how societies dispose of what remains, but what kinds of life emerge around its continued presence.

x

Climate transition is often presented as a project of collective planetary repair. Net zero targets, renewable energy infrastructures, adaptation strategies and carbon removal schemes all point toward a common objective, securing a more stable climatic future. Over the past half century, advances in Earth system science, climate modelling and global environmental governance have transformed climate from an object of observation into an object of management. The planet became something that could be measured, predicted and, crucially, steered.

Yet stability is not being distributed equally.

Across the world, climate transition is reorganising where environmental risk is located, who is expected to endure it, and which forms of exposure become politically acceptable. New infrastructures are being built in the name of decarbonisation. New territories are being incorporated into the machinery of climate governance. But the benefits and burdens of these transformations do not fall evenly. As some spaces become increasingly protected, others are asked to absorb the material, environmental and social costs of securing those futures.

This is where carbon afterlives become visible. Looking beyond emissions alone reveals a different set of questions. If earlier carbon regimes organised territories around extraction, combustion and accumulation, contemporary climate regimes are increasingly organised around differential protection. The question shifts from carbon emissions to how future survivability is allocated. Whose environments are secured? Which infrastructures are protected? Which territories become expendable? And what kinds of worlds are being assembled through the pursuit of climate stability?

The Geography of Tolerable Instability

One of the defining characteristics of contemporary climate transition is the emergence of what might be called managed exposure zones. These are territories whose instability is not treated as a failure of climate governance but as a condition of its operation.

Across the world, new geographies are emerging around lithium frontiers, desalination corridors, carbon storage basins, renewable energy zones and critical mineral extraction regions. These spaces are frequently presented as necessary components of decarbonisation, yet their inclusion within green transition often depends precisely upon their capacity to absorb environmental disruption, infrastructural burden and social risk.

Saudi Arabia’s NEOM is one such example. Often framed as a model for sustainable urbanism powered by renewable energy and advanced digital infrastructure, the project is typically discussed through the spectacular imagery of The Line and other futuristic developments such as Trojena. Viewed through the lens of carbon afterlives, however, NEOM can also be understood as a territorial reorganisation of exposure. The project requires vast energy infrastructures, desalination systems, logistics networks, construction materials and continuous environmental management across an arid landscape already characterised by ecological vulnerability.

Here, stability is less a realised condition than a governing promise. The future urban order projected by NEOM mobilises enormous investments, infrastructures and territorial interventions in the present, regardless of whether the city ever materialises in its intended form. The highly controlled environments promised by the project depend upon extensive systems operating beyond its boundaries—systems that absorb environmental, material and political pressures elsewhere.

What emerges is not simply a sustainable city but a geography in which resilience is concentrated while instability is redistributed. The transition being enacted is not from carbon to post-carbon, but from one arrangement of exposure to another. Climate governance increasingly secures futures through the strategic management of uncertainty in other places. The afterlife of carbon therefore lies not only in the redistribution of environmental exposure, but also in the proliferation of speculative climate futures whose material consequences arrive long before the futures themselves.

Climate Adaptation as Zoning

The field of climate adaptation is understood as a universal response to shared climatic threats. In practice, adaptation rarely arrives everywhere at the same speed, scale or intensity. Through this model, “resilience” is assembled via investments in infrastructure, governance, technology and maintenance, and these capacities are distributed unevenly across territories.

As a result, adaptation increasingly takes on a spatial dimension. Certain districts become sites of concentrated investment in cooling systems, flood defences, energy security and digital monitoring, while other places remain more exposed to heat stress, infrastructure failure and environmental risk. The question is therefore not simply whether adaptation occurs, but how resilience becomes attached to particular spaces and material arrangements.

From the perspective of carbon afterlives, what matters is not only the technical performance of adaptive infrastructures but the kinds of worlds they produce. Increasingly, climate resilience is being imagined through enclosed, highly managed environments in which atmospheric conditions become objects of continuous technological control.

Saudi Arabia’s Mukaab offers a revealing example. Often discussed as an architectural spectacle, the project is perhaps most interesting as an experiment in environmental management. AI-powered mobility systems, climate-controlled interiors and immersive digital environments are designed to organise movement, circulation and experience within an engineered atmosphere. As I argued elsewhere, the most interesting feature of the Mukaab may not be the cube but its elevators. These systems transform mobility into climate infrastructure, organising bodies, energy, logistics and perception inside a carefully managed interior world.

The significance of projects such as the Mukaab lies in the futures they project. Climate adaptation increasingly takes the form of environmental engineering, creating controlled alternatives to existing climatic conditions. Atmosphere, circulation and sensory experience become objects of continuous calibration, coordinated through dense networks of digital and material infrastructure. Urban space is reimagined as an operational environment in which stability is actively produced rather than simply inherited from surrounding ecological conditions.

A classed distinction is becoming visible within climate adaptation. For vulnerable populations, adaptation is typically framed in terms of resilience, securing water supplies, protecting homes, sustaining livelihoods and maintaining ways of life under increasingly difficult environmental conditions. The objective is endurance. For wealthier populations, however, adaptation increasingly takes the form of environmental redesign. Projects such as the Mukaab promise alternative climatic realities in which temperature, atmosphere, mobility and sensory experience are engineered through technological systems. The question is no longer how to live with climate variability, but how to selectively escape it.

From a digital theory perspective, this signals a broader transformation in the object of governance. Attention shifts from the management of territory to the management of conditions. Climate becomes an operational variable, rendered legible through sensors, modelled through algorithms and adjusted through infrastructural feedback loops. The city emerges as a programmable interface through which environmental stability is continuously produced, maintained and optimised.

Viewed through the lens of carbon afterlives, the political significance of these projects extends beyond architecture or urban development. What emerges is a geography in which resilience is unevenly distributed and exposure differentially managed. Some populations are asked to adapt to climatic uncertainty, while others gain access to increasingly controlled environments designed to insulate them from it. Climate governance thus becomes concerned not only with reducing emissions or managing risk, but with organising who lives with instability and who is able to inhabit its engineered alternatives.

Uneven Survivability

Climate discourse often gravitates toward the language of catastrophe. Yet one of the defining features of climate change may be the increasingly uneven distribution of survivability. Carbon afterlives draw attention to the infrastructures, institutions and forms of governance through which populations acquire different capacities to endure environmental change. Insurance, cooling, mobility, relocation and protective infrastructure increasingly shape how climate risk is experienced, managed and absorbed.

Florida offers a revealing example. Despite growing awareness of sea-level rise and flood risk, the state’s population increased by more than 3 million people between 2010 and 2023. At the same time, insurers have withdrawn from some of the most exposed markets, premiums have risen sharply, and flood-related losses continue to mount. These developments suggest that climate exposure alone does not determine where people live. The capacity to remain in place increasingly depends upon access to insurance, public investment, resilient infrastructure and financial resources capable of absorbing risk.

What emerges is a changing geography of protection. Environmental hazards remain broadly distributed, while the mechanisms that buffer their effects become increasingly uneven. Some communities gain access to the infrastructures, services and investments that make environmental uncertainty more manageable. Others encounter rising costs, diminishing protections and growing exposure. Climate adaptation therefore unfolds not only through the management of environmental risk, but through the allocation of the resources required to live with it.

Carbon afterlives reveal how climate governance is increasingly organised around the distribution of survivability. The question extends beyond how societies respond to environmental change. It concerns how protection, security and endurance are arranged across populations, territories and infrastructures, and how those arrangements shape the futures that remain possible.

Luxury Resilience

Perhaps nowhere is this differentiation more visible than in the emergence of luxury resilience.

Climate adaptation increasingly appears as a premium condition. Climate-proof enclaves, AI-managed districts, highly serviced urban interiors, private fire protection systems and technologically insulated developments point toward a future in which resilience becomes a scarce and valuable resource.

Projects such as NEOM, The Line and other Gulf mega-developments are often presented as visions of future urban life. Through the lens of carbon afterlives, they can also be understood as projects for reorganising environmental security. Comfort, cooling, mobility and infrastructural reliability become central design objectives, shaping highly managed environments in which atmospheric conditions are continuously maintained through technological systems. These developments project futures organised around differentiated access to stability, concentrating protection and environmental control within carefully engineered spaces.

The crucial question concerns the worlds emerging through climate transition. Carbon afterlives draw attention to the social, political and spatial arrangements being produced alongside decarbonisation. New geographies of protection and exposure are taking shape. New infrastructures of endurance are being built. New forms of political belonging are emerging around access to security, mobility and environmental stability. While the atmosphere remains shared, the conditions through which people experience and endure climatic change are becoming increasingly differentiated.

Climate transition extends beyond the reduction of emissions. It is also a process of reorganising how protection, risk and opportunity are distributed across populations and territories. Carbon afterlives reveal how future worlds are being assembled through decisions about infrastructure, adaptation, investment and governance.

These decisions shape who remains protected, who carries increasing exposure, and which futures become possible.

x

Mountains for the Atmosphere

One of the most striking developments in today’s climate politics is unfolding in the southeastern coast of the Arabian Peninsula.

Rising across northern Oman, the Hajar Mountains contain some of the largest exposed formations of peridotite in the world. These ancient rocks, pushed to the Earth’s surface through tectonic upheaval millions of years ago, are attracting attention for their unusual capacity to transform atmospheric carbon dioxide into stone.

Beneath parts of the mountains lie vast formations of peridotite, a dense ultramafic rock capable of reacting with atmospheric carbon dioxide and gradually transforming it into solid carbonate minerals. Researchers and private companies are exploring whether these naturally occurring processes can be accelerated through carbon mineralisation technologies, potentially turning geological formations into large-scale systems for atmospheric carbon removal.

What makes this project so compelling is that it reveals how climate change is currently being imagined and governed. Mountains, subsurface geology and mineral formations are no longer treated purely as landscapes, resources or sites of extraction. They are being repositioned as infrastructures capable of stabilising atmospheric futures under conditions of continuing climatic instability.

The implications are difficult to overstate. Geological formations created millions of years ago are being reimagined as mechanisms for governing atmospheric conditions that now appear difficult to manage through emissions reductions alone. Geology is becoming enrolled within systems of climate management.

Similar developments are emerging elsewhere. Across the North Sea, governments and energy companies are converting depleted offshore oil and gas reservoirs into long-duration carbon storage basins. In Iceland, direct air capture facilities remove carbon dioxide from the atmosphere before injecting it into basalt formations where it mineralises into stone. Throughout North America, proposals for extensive carbon pipeline networks continue to reshape regional politics as captured carbon is transported across vast distances towards geological storage sites. Increasingly, landscapes, infrastructures and subsurface formations are being reorganised around the challenge of governing carbon after its release.

The Rise of Carbon Management

The language of decarbonisation or energy transition, although important, doesn’t fully explain what is happening. A parallel transformation is taking place within approaches to climate change. Carbon is being treated less as a pollutant and more as a material resource whose movement, storage, containment and long-term fate must be actively managed for millennia to come.

For much of the past two decades, climate politics focused heavily on emissions reduction and behavioural transformation. Citizens were encouraged to recycle, consume differently, travel less and reorganise everyday life around lower-carbon forms of conduct. Responsibility appeared closely tied to the reduction of emissions at their source. Climate governance was therefore imagined largely through the language of prevention.

What is emerging today operates according to a somewhat different logic. Atmospheric carbon remains a problem, but it is accompanied by an expanding set of technical, geological and institutional systems designed to manage carbon after its release. Captured carbon becomes economically valuable. Stored carbon becomes infrastructural. Mineralised carbon acquires the political significance of permanence. The centre of gravity shifts gradually from reducing emissions alone towards governing carbon across extended temporal and spatial scales.

This shift is unevenly distributed across the world and reveals important differences in how climatic futures are being imagined. European climate discourse continues to be shaped by narratives of transition, reduction and decarbonisation. Within these frameworks, climate action is frequently understood as a movement away from fossil dependence towards alternative energy systems capable of sustaining economic and social life under lower-carbon conditions.

Elsewhere, particularly within parts of the Gulf, climate governance is organised around questions of carbon management, removal and storage. Hydrocarbon production remains central to economic life, yet this does not necessarily place such states outside contemporary climate politics. Instead, they are positioning themselves within emerging systems of atmospheric management designed to coexist with ongoing extraction. Carbon capture, geological storage and mineralisation technologies enable hydrocarbons and climate governance to become entangled within the same infrastructural landscape.

From Contamination to Stewardship

What interests me about this transformation is that it recalls a much older set of questions that have long occupied environmental thought.

For decades, pollution studies, environmental justice scholarship and critical ecological research have traced the movement of contaminants through bodies, landscapes and ecosystems. Their central concern was often persistence. Waste rarely disappeared when it left the site of its production. It accumulated, circulated, settled and returned, often producing consequences that were distributed unevenly across communities and generations. Environmental politics therefore involved learning how to trace the afterlives of substances long after their apparent disappearance.

Carbon invites a similar mode of attention.

Although carbon dioxide differs from many conventional pollutants, contemporary climate governance is transforming it into an object that generates its own infrastructures of monitoring, containment and long-term responsibility. Carbon is measured, modelled, transported, verified, offset, stored and fixed within geological formations. Entire industries are emerging around its management. New forms of expertise are developing around its monitoring. Vast infrastructures are being constructed around its containment.

What becomes visible through these developments is a subtle but significant shift in the environmental imagination.

Pollution discourse has often begun with contamination and exposure. It has asked how harmful substances move through environments and how their effects become unevenly distributed. Contemporary carbon governance directs attention towards another set of questions concerning maintenance, responsibility and long-duration care. Once carbon is stored, who monitors it? When geological formations become repositories for atmospheric carbon, who assumes responsibility for their integrity across decades or centuries? What institutions are required to maintain systems whose intended lifespan extends far beyond ordinary political cycles?

These questions suggest that climate governance is gradually becoming organised around forms of stewardship that would have been difficult to imagine only a generation ago. Carbon stores require monitoring. Adaptation infrastructures require maintenance. Environmental archives require preservation. Geological repositories require ongoing oversight. The management of atmospheric futures depends upon systems of care, observation and responsibility that extend across generations.

There is a curious contrast here with other traditions of stewardship. Across many cultures, care for land has been grounded in accumulated knowledge passed between generations, rooted in close relationships with certain landscapes and an understanding of their seasonal rhythms and histories. Carbon stewardship operates differently. Its objects are often invisible, its infrastructures buried underground or offshore, and its timescales measured in centuries rather than lifetimes. The obligation to care persists, but it is increasingly directed towards atmospheric futures known through scientific instruments, geological models and monitoring systems rather than through inherited relationships with place. At the same time, these forms of stewardship are deeply entangled with new opportunities for investment, accumulation and profit. Carbon stores, removal technologies and monitoring systems are becoming economic assets as well as environmental responsibilities, creating situations in which care and capital become increasingly difficult to separate.

Seen from this perspective, the significance of projects such as those emerging in Oman extends well beyond carbon removal. They reveal an important transformation in how societies relate to environmental futures. The challenge is no longer confined to reducing emissions, important though that remains. It involves constructing institutions, infrastructures and forms of responsibility capable of managing the enduring consequences of industrial carbon circulation.

The question that follows is therefore not simply where carbon goes.

It is what kinds of political, social and material worlds emerge around the effort to make it stay there.

x

Perhaps the most important shift in climate politics is not the return of denial. Nor is it the enduring influence of fossil fuel interests, although both remain powerful.

The deeper shift is something quieter. It sits beneath election cycles, policy announcements and the increasingly familiar arguments over net zero. It is the growing recognition that climate instability may no longer be fully preventable.

You can see this recognition appearing in different places at once. In governments that continue to talk about climate ambition while quietly diluting commitments. In insurers retreating from regions increasingly exposed to wildfire, flooding and extreme heat. In policy circles where conversations that once focused on emissions reductions now spend as much time discussing resilience, adaptation and risk management.

For much of the last decade, net zero functioned as a political horizon. It offered a way of imagining that decarbonisation could be reconciled with economic growth, geopolitical stability and democratic consent. The promise wasn’t ever only environmental, it was always also institutional. Climate change could be addressed without fundamentally reorganising the systems that had produced it.

That horizon has not disappeared entirely, but it is becoming harder to sustain.

Last week, former UK Prime Minister Tony Blair argued that existing approaches to net zero are becoming politically and economically unsustainable. The backlash was immediate and predictable. Still, the significance of Blair’s intervention lies less in the argument than in its timing. Similar recalibrations are appearing across policy circles, energy markets, security institutions and sections of the climate establishment.

Taken together, they point towards a subtle but important shift. The organising question is no longer how to prevent climate disruption. Increasingly, it is how to manage a future in which disruption is expected.

This shift can be observed almost everywhere. Throughout Europe, farmers protest environmental regulations while governments quietly dilute climate commitments. In Washington, the Inflation Reduction Act has tied decarbonisation to industrial competition and strategic rivalry with China. Across insurance markets, entire regions are becoming difficult to insure against wildfire, flooding and extreme heat. Meanwhile, attention is increasingly directed not only towards renewable energy, but towards adaptation finance, climate resilience infrastructure, climate-risk analytics and geoengineering research.

The transition has become geopolitically uneven. China now dominates much of the industrial infrastructure underpinning decarbonisation — from solar panels and batteries to critical mineral processing and electric vehicles. Increasingly, Western governments face a dilemma they rarely articulate openly, how to accelerate the energy transition without deepening dependence on Chinese manufacturing capacity.

The politics of climate change is slowly being reorganised around a different assumption. Stability can no longer be restored so, now, instability must be governed.

The implications of that shift are profound.

What will follow is a politics less concerned with restoring equilibrium than with administering disequilibrium.

Governing the climate increasingly recasts the atmosphere as an object of calculation, monitoring and intervention. Satellites track emissions in real time. Carbon border mechanisms transform atmospheric accounting into trade policy. Earth observation systems feed predictive models designed to anticipate ecological disruption before it arrives. Even solar geoengineering — once confined to the margins of scientific speculation — now circulates within policy reports, philanthropic funding networks and national security discussions.

The Politics of Uneven Survival

Climate realism does not signal the abandonment of climate governance. If anything, it reflects a growing recognition that climate governance is entering a new phase. For much of the last decade, climate politics was organised around the possibility of avoiding disruption. The language was one of transition, transformation and prevention. Today, a different vocabulary is beginning to take hold. This language speaks of adaptation, resilience, risk and strategic protection.

Behind these terms sits an uncomfortable reality. Climate change is now sufficiently advanced that many institutions are planning not only for mitigation, but for a future in which environmental instability becomes a permanent feature of political and economic life. This changes the nature of the challenge. The ambition is no longer to eliminate instability altogether. It is to determine what can still be stabilised within it.

That shift has profound consequences. Some territories will be protected through investment, cooling infrastructure, insurance mechanisms and technological adaptation. Others will face escalating heat, deteriorating infrastructure, extraction pressures or declining access to protection. The uneven geography of climate change is increasingly being mirrored by an uneven geography of resilience.

We are already beginning to see this logic emerge. Across parts of the world, affluent communities are investing in private forms of adaptation, from climate-proof housing and water security to sophisticated risk modelling and disaster planning. At the same time, many of the communities most exposed to climate disruption remain the least equipped to absorb its consequences. The gap between exposure and protection is becoming a political question in its own right.

What is striking is that this recalibration is no longer confined to openly conservative or fossil-fuel aligned actors. Variations of it are appearing across parts of the climate establishment. Bill Gates’ emphasis on adaptation technologies, advanced nuclear energy, carbon removal and geoengineering reflects more than technological optimism. It points towards a growing recognition that emissions trajectories, energy demand and geopolitical fragmentation may prove far more resistant to rapid reversal than earlier transition narratives assumed. The language remains hopeful, but the underlying assumptions are changing.

Climate change is increasingly framed less as a problem to be solved and more as a condition to be managed. The emphasis shifts towards adaptation, prioritisation and strategic intervention. Under these conditions, difficult questions begin to surface.

Which infrastructures are considered essential enough to defend?

Which industries remain politically and economically indispensable?

Which communities receive protection from heat, energy volatility, insurance withdrawal or resource scarcity, and which are expected to absorb greater levels of risk? These are not technical questions. They are political ones. The danger is that they respond in ways that normalise unequal exposure to its consequences.

Overshoot—the increasingly plausible breaching of warming thresholds once presented as safeguards—becomes something to manage rather than avoid. Adaptation becomes less about collective security and more about negotiating who bears the costs of instability. In that sense, climate realism is not merely a shift in environmental policy. It is the emergence of a new politics of survival.

And survival, as history repeatedly reminds us, is rarely distributed equally.

The Atmosphere Becomes Infrastructure

We are entering a period defined less by ecological prevention than by continuous planetary management: atmospheric sensing, predictive adaptation, climate modelling, Earth observation systems, AI-assisted forecasting and expanding discussions around geoengineering. From wildfire prediction systems in California to satellite-based carbon surveillance and AI-driven climate-risk modelling for insurers, climate governance increasingly operates through continuous monitoring and real-time adaptation.

Net zero begins to resemble cybernetics at planetary scale — a permanent infrastructural effort to monitor, calibrate and manage instability rather than transcend it. This is not the end of climate politics but the end of the green consensus. The political question of the coming decades will no longer be whether governments act on climate change, but whose infrastructures, territories and futures climate governance is ultimately designed to secure — and who becomes expendable within its calculations.

Who Governs the Sky?

If the green consensus is ending, what replaces it?

One possibility is already emerging, a fragmented politics of atmospheric intervention. As faith in coordinated decarbonisation weakens, interest in solar radiation management, weather modification and geoengineering continues to grow. What was once treated as speculative science fiction increasingly enters mainstream policy discussion, philanthropic funding networks and national security discourse. The atmosphere is no longer thought of in policy and law as a shared ecological condition, but as an increasingly governable and contested strategic domain. Planetary governance is not abandoned but escalates.

Recent debates around solar geoengineering reveal how quickly the language of emergency can legitimise extraordinary intervention. Proposals to dim sunlight, alter atmospheric reflectivity or artificially suppress warming are increasingly framed as pragmatic responses to political failure. Nevertheless, these interventions raise profound geopolitical questions. Who controls the thermostat of the planet? Which regions bear the risks of altered rainfall patterns, agricultural disruption or hydrological instability? Who decides what level of warming is tolerable — and for whom?

Already, accusations of atmospheric manipulation and “cloud stealing” circulate within drought-stricken regions experiencing severe water stress. Climate instability increasingly intersects with nationalism, territorial anxiety and resource insecurity. Under conditions of intensifying crisis, the atmosphere risks becoming a site of geopolitical suspicion and conflict. At the same time, private actors and start-ups are beginning to pursue geoengineering experiments with astonishing speed, often ahead of meaningful international governance structures.

The danger is not only technological overreach. It is the collapse of the multilateral imagination.

Progressively, multilateral climate governance is dismissed by elites as politically naïve or economically unrealistic. In its place emerges a fragmented landscape of strategic adaptation, national resilience planning and experimental interventionism. Under conditions of permanent instability, exceptional measures become easier to justify. This may be the endpoint of climate realism, not collective planetary stewardship, but competitive atmospheric management.

The atmosphere becomes territorial once again — not only through borders, carbon accounting and strategic supply chains, but through direct intervention into planetary systems themselves. What disappears in this transition is the assumption that climate politics was ever fundamentally about equality. The future now being organised is not one in which everyone survives equally. It is one in which survivability becomes administered, unevenly distributed and increasingly securitised.

For much of the last decade, climate politics was organised around the promise of a shared future. What is now emerging is something different. A politics less concerned with preventing instability than with governing it. Less concerned with planetary repair than with the selective preservation of strategic systems. The question is no longer whether humanity can stabilise the climate. It is who acquires the authority to manage instability — and who is asked to live with the consequences.

That may prove to be the defining political struggles of the twenty-first century.

x

Upgrade Culture

A few months ago, my phone contract ended.

The battery barely lasts half a day now. Apps freeze unpredictably. It overheats while charging. Increasingly, it feels less like a device naturally ageing and more like something slowly becoming incompatible with the world around it.

Each time this cycle comes around, I tell myself I should buy a Fairphone instead.

Repairable. Longer lasting. Replaceable components. A phone designed against disposability rather than around it.

Then I look at the upfront cost and quietly close the tab.

This is usually where discussions about climate responsibility become moralised. Consumers should buy better things. Repair more carefully. Upgrade less. Recycle properly.

But the longer I think about climate infrastructures, the harder it becomes to see these decisions as simply individual choices.

The issue is not that sustainable alternatives do not exist.

It is that entire systems are organised around making other choices easier, cheaper and more normal.

Most phones are not designed to last. Contracts distribute costs in ways that make replacement feel frictionless while repair feels expensive. Batteries are sealed. Screens are difficult to replace. Software updates gradually render older devices slower and less compatible. Cloud storage, app ecosystems and upgrade cycles quietly encourage continuity within the same technological system.

The result is not simply waste.

It is persistence.

A phone rarely disappears when we stop using it.

Its materials continue moving through mining systems, shipping routes, data centres, electricity grids, recycling infrastructures and waste economies distributed across multiple continents. The device sitting in your hand is connected to copper extraction in Zambia, lithium mining in Chile, semiconductor manufacturing in Taiwan and rapidly expanding digital infrastructures elsewhere.

What appears lightweight and immaterial at the point of use depends upon extraordinarily heavy systems underneath.

The Other 1%

A few years ago, I attended an online talk by waste scholar Josh Lepawsky. One point has stayed with me ever since.

Lepawsky argued that we often confuse scale with magnitude.

The biggest visible pile of waste is not always where the most consequential intervention lies.

He gave the example of lithium extraction, a relatively small reduction (1%!) in extraction at the mining stage could produce disproportionately larger reductions further along the supply chain. Intervening earlier in infrastructural systems can reshape everything downstream.

Lepawsky’s work is useful here because he argues that “e-waste” is not a fixed category but something continuously organised through systems of collection, classification and reporting.

The image looked like this:

The three lines beneath the EU symbol refer to the administrative stages through which electronic waste becomes governable: collection, treatment/recovery and reporting/compliance. The point of the diagram is not simply that waste gets recycled, but that institutions attempt to make discarded electronics measurable and manageable through these systems.

But as Lepawsky suggests, electronic objects constantly exceed those categories. The same laptop or phone may move between being classified as waste, repair material, reusable equipment or commodity depending on where it circulates and who is handling it.

That idea increasingly feels important for understanding climate politics more broadly.

Much contemporary climate governance focuses on visible endpoints of consumption, recycling bins, electric cars, efficient homes, individual carbon footprints, whether someone upgrades their phone too often.

But focusing only on the final object can obscure the infrastructures that made that object inevitable in the first place.

This is partly what scholars mean by carbon lock-in.

Carbon-intensive systems persist not simply because fossil fuels exist, but because entire infrastructures, institutions and habits become organised around them over time. Roads shape transport systems. Housing stock shapes heating systems. Data centres shape digital life. Contracts shape consumer behaviour. Supply chains shape what kinds of choices become economically realistic in the first place.

Emissions become built into the ordinary organisation of everyday life.

Polite Littering

Perhaps this is also why waste matters politically.

Waste is not simply what gets thrown away. It reveals what systems are willing to leave unresolved.

Waste scholar Randa Kachef uses the phrase “polite littering” to describe forms of disposal that preserve the appearance of cleanliness while relocating waste elsewhere. The litter is not eliminated so much as displaced beyond the immediate field of visibility.

Modern waste systems often operate this way.

The UK exports plastic waste abroad despite domestic recycling targets. Electronic waste is routinely shipped across borders under the language of recovery and recycling. Even within cities, waste infrastructures are unevenly distributed, some neighbourhoods remain physically closer to incinerators, flood zones, roads and industrial pollution than others.

The residue remains.

Distance simply reorganises who lives alongside it.

Perhaps contemporary climate governance increasingly operates this way too.

In the UK, Carbon Capture and Storage projects such as HyNet NorthWest and the East Coast Cluster are presented as central to industrial decarbonisation strategy. These infrastructures aim to capture emissions from refineries, power stations and heavy industry before transporting carbon dioxide through pipeline systems for long-term storage beneath the North Sea.

Supporters argue these systems are necessary for reducing emissions in sectors where electrification remains difficult.

But materially, carbon is not removed from industrial systems altogether.

It is compressed, transported, buried underground and monitored indefinitely through new infrastructures of storage and verification. Existing fossil fuel infrastructures — pipelines, offshore engineering systems and depleted gas fields — are repurposed into systems for long-term carbon management.

The carbon remains.

What changes is where it goes, how it circulates and who lives alongside the infrastructures required to contain it.

Uneven Persistence

What interests me about Carbon Capture and Storage is not simply the technology itself, but the kinds of futures it makes economically valuable.

Projects such as HyNet NorthWest are framed not only as climate interventions, but as opportunities for industrial continuity, regional regeneration and investment. Existing oil and gas infrastructures become repurposed into new systems of carbon management. Carbon storage becomes economically productive precisely because carbon continues moving through infrastructures of transport, monitoring and containment.

Persistence becomes investable.

But not all forms of carbon persistence are valued equally.

This becomes visible if you look somewhere like Zambia.

After independence, Zambia largely pursued hydropower rather than large-scale fossil fuel development. Compared to many industrial economies, its electricity system remained relatively low-carbon for decades.

But that pathway did not produce the same forms of industrial accumulation associated with fossil-fuelled development elsewhere. Instead, Zambia became increasingly dependent upon climate-sensitive hydro infrastructures shaped by colonial extraction, debt dependency and uneven development.

Now climate change is destabilising the very infrastructures that once appeared comparatively sustainable.

As drought conditions intensify across southern Africa, water levels at the Kariba reservoir have fallen dramatically, contributing to electricity shortages and energy instability. A country that contributed relatively little to global emissions now faces escalating climate vulnerability precisely through the infrastructures that once enabled lower-carbon development.

At the same time, Zambia is becoming increasingly integrated into global green transitions as a site of copper extraction for electric vehicles, batteries and renewable infrastructures elsewhere.

The materials continue moving.

What emerges is a profoundly uneven geography of persistence.

In the UK, carbon persistence becomes investable through systems like CCS. Existing fossil infrastructures are reorganised into profitable systems of storage, monitoring and long-term management.

In Zambia, restraint and lower historical emissions do not produce equivalent forms of investment or security. Instead, low-carbon development becomes entangled with drought, extraction, debt and infrastructural vulnerability.

One form of persistence is subsidised.

Another is abandoned to climate instability.

Replacement

This is why the problem is no longer simply whether transition happens.

The question is what kinds of persistence contemporary economies know how to value.

Keeping an existing phone for another five years generates relatively little economic value. Maintenance slows circulation. Durability interrupts replacement. Restraint limits accumulation.

Obsolescence keeps materials moving.

Perhaps this is one of the deeper contradictions within green capitalism.

Systems organised around continual circulation struggle to value forms of maintenance, durability or restraint because containment without movement produces limited opportunities for accumulation.

The easiest thing for contemporary economies to manage is not disappearance, but replacement.

x

With thanks to Matt Lane for conversations and shared thinking around infrastructure, climate and carbon afterlives.

Last week, I started with a simple question, where does carbon go?

Not in the abstract sense of emissions targets or global totals, but materially, once it leaves the point where it is produced. What happens to it afterwards, and how does it continue to move through the systems designed to manage it.

One answer is relocation.

A heat pump is supposed to feel local.

It sits quietly outside a house, replacing a boiler. Governments subsidise them. Energy companies advertise them as part of a cleaner future. Installed correctly, they can reduce household emissions significantly.

But a heat pump is not only a domestic appliance.

Its existence depends on supply chains stretching across mines, factories, shipping routes, electricity grids, rare minerals and rapidly expanding digital infrastructure. The transition to electric heating does not simply reduce carbon. It reorganises where energy demand, extraction and industrial activity happen.

Somewhere between a suburban home in Britain and a data centre in India is the story of how contemporary decarbonisation actually works.

Last month, The Guardian reported that officials had significantly underestimated the future carbon impact of AI-linked data centres in the UK.

At roughly the same time, the UK government continued accelerating planning reforms designed to fast-track new data infrastructure, including treating large facilities as Critical National Infrastructure and streamlining approvals through the Nationally Significant Infrastructure Projects regime.

Meanwhile, in Hyderabad — one of India’s fastest-growing data infrastructure hubs — operators and climate researchers were warning that rising temperatures and worsening water scarcity were placing increasing pressure on cooling systems needed to keep facilities operational.

These stories might initially seem unrelated,
AI growth in Britain, planning reform, heatwaves in India.

But they are connected by a larger question that sits quietly beneath contemporary climate politics

What happens to carbon after extraction, combustion and “offsetting”?

For the last decade, net zero has largely been imagined through disappearance. Fossil fuels are phased out. Homes are electrified. Cars become battery-powered. Energy becomes renewable. Everyday life for some appears cleaner, quieter and more efficient.

And in many ways, it is.

Heat pumps can reduce dependence on gas heating. Electrification can reduce direct emissions and improve urban air quality. Renewable energy remains essential to limiting planetary warming.

But the longer you sit with these transitions, the more another reality begins to emerge.

Carbon rarely disappears completely unless it is never produced in the first place.

Instead, it is increasingly relocated.

Into electricity grids, mineral extraction, logistics systems, cooling infrastructures, battery supply chains, server halls and carbon accounting systems. Emissions become displaced geographically and infrastructurally rather than fully eliminated outright.

The apparent purity of low-carbon systems often depends upon forms of opacity elsewhere.

Clean at the Point of Use

This becomes especially visible through data centres.

Over the last two years, the UK has rapidly expanded support for large-scale digital infrastructure as part of its future growth strategy. Facilities linked to AI, cloud computing and digital services are increasingly treated as strategic infrastructure necessary for economic competitiveness and national resilience. Yet data centres are also among the most energy-intensive buildings ever constructed, requiring enormous and continuous supplies of electricity alongside extensive cooling systems, backup generators and expanding grid infrastructure.

Data centres are not secondary to net zero transitions.

They are increasingly one of the conditions that sustain them.

A heat pump installed in a newly retrofitted London home may appear locally clean, no gas boiler, lower direct emissions and improved efficiency ratings. Yet electrified homes do not operate in isolation. As electricity systems become increasingly digitised, they rely more heavily on smart meters, automated grid-balancing platforms, cloud-based energy management systems and expanding computational infrastructure. The carbon intensity of these systems is not fixed. In some regions they are increasingly powered by renewables; in others they remain tied to fossil-fuelled grids and energy-intensive cooling infrastructures. What appears locally decarbonised can therefore still depend upon wider infrastructural systems whose environmental burdens are distributed unevenly across different places.

The point of use becomes cleaner while the wider infrastructural system expands.

This redistribution is often difficult to perceive because transition systems increasingly operate through distance and abstraction. Emissions shift into supply chains, outsourced manufacturing systems, electricity infrastructures and computational systems largely removed from everyday visibility.

What appears as decarbonisation in one place may therefore depend upon intensified infrastructural demand elsewhere.

Hyderabad

This becomes particularly visible in Hyderabad.

Over the last decade, Hyderabad has emerged as one of India’s largest data infrastructure corridors, attracting major investment from companies including Amazon Web Services, Microsoft and Google. At the same time, these facilities are increasingly exposed to escalating climate instability.

As of 2026, Hyderabad is already experiencing recurrent extreme heat events exceeding 40°C during summer months, alongside intensifying water stress across the wider Telangana region. According to projections from the IPCC and regional climate assessments, extreme heat events across parts of southern India are expected to become more frequent, prolonged and severe over the coming decade, while wet-bulb temperatures — combinations of heat and humidity dangerous for both human labour and technical infrastructure — continue rising.

For data centres, this creates a profound contradiction.

Server systems require tightly stabilised thermal conditions to function reliably. As temperatures rise externally, cooling systems must consume substantially more electricity simply to maintain operational stability. Water demand for cooling may intensify precisely as regional water scarcity worsens. Extreme heat simultaneously increases the likelihood of grid instability, outages and equipment failure.

The hotter the climate becomes, the more energy-intensive cooling infrastructures must become in response.

Climate change therefore increasingly destabilises the very infrastructures required to sustain digitally managed net zero systems.

When Infrastructure Fails

This tension extends beyond climate into geopolitics and security.

Earlier this year, drone strikes linked to escalating regional conflict in the UAE disrupted power systems connected to AWS-linked infrastructure facilities in the Gulf region, exposing the territorial vulnerability of supposedly seamless digital systems. Suddenly, what appeared abstract became materially concentrated again, substations, transmission systems, cooling infrastructure, backup generators and logistical corridors vulnerable to interruption and physical attack.

Infrastructure often becomes most visible at the moment of failure.

These examples reveal something broader about contemporary climate governance.

Net zero transitions are frequently imagined through narratives of purity – clean growth, clean mobility, clean homes, clean technology. Yet many of these systems remain deeply interdependent with expanding infrastructures of extraction, energy demand, cooling, logistics and computation.

Decarbonisation does not necessarily unfold through dematerialisation.

It often unfolds through infrastructural expansion.

Relocation & Opacity

This is why relocation matters politically.

Some populations experience cleaner air, electrified homes and digitally optimised systems. Others absorb intensified extraction, industrial expansion, water stress, climatic instability and infrastructural burden. Carbon moves through atmospheres, oceans, soils, server infrastructures, financial systems and supply chains unevenly across different territories and generations.

The issue is not whether transition should occur. Emissions reductions remain essential.

The issue is how transitions are organised materially, whose infrastructures sustain them and who lives alongside their afterlives.

Carbon afterlives are therefore not only atmospheric.

They are infrastructural, ecological and geopolitical.

What matters is not simply where carbon is emitted, but how societies redistribute responsibility for living with carbon’s persistence across different places, bodies, infrastructures and futures.

x

With thanks to Padmini Ray Murray, whose insights and conversation helped shape many of the questions explored in this piece.

This is the first post in a short series I’ll be writing over the next eight weeks.

I’ve been developing an idea that sits across my practice, dissertation and a longer book project, and I want to begin opening it up here as a way of working through it in public. Each post will stay with one aspect — turning it over, testing it and seeing what it makes visible.

I’m calling this idea carbon afterlives.

Most of us have learned to imagine climate change through numbers.

A flight emits a certain amount of carbon. A house can become “net zero.” A company can reduce its footprint. An electric car is cleaner than a petrol one. Emissions go up, then — through technology, policy and better choices — they begin to come down again.

Carbon appears as something that can be measured, balanced and eventually removed from the world.

But where, exactly, does it go?

When storage fails

In 2020, a wildfire burned through parts of California’s Sierra Nevada. Some of the land it passed through had already been counted.

The forests were part of a carbon offset scheme. Their trees had been drawn into a system that tracks carbon across space. Growth had been estimated, futures projected and, from that, a quantity established — how much carbon this forest might hold over time.

That future was translated into credits.

Those credits had already been used. Somewhere else, emissions had been released — through industry, transport and everyday activity — and accounted for against this forest. Not removed, but balanced.

The logic is familiar enough. What is produced here can be handled there.

It now shapes everyday life, flights are offset, products become “carbon neutral,” companies promise net zero while continuing to emit.

It is a system that depends on calculation. On the idea that different places, different times and even different forms of life can be made equivalent through measurement. That a tonne of carbon released now can be set against a tonne stored elsewhere, later.

In some contexts, these arrangements bring investment into landscapes that might otherwise be under pressure from extraction or development. Forests are protected, sometimes expanded. New forms of income appear.

In others, they reorganise land in more difficult ways — tightening control, shifting access and asking certain places to take on the work of storage for emissions produced somewhere else.

Often, it is both at once.

On paper, though, the outcome is clear. The carbon has been dealt with.

Then the fire came through.

Carbon held in trees and soil over decades returned to the atmosphere in a matter of days. What had been stabilised through models and accounting became unstable again.

Nothing new had been produced. No new decision had been made.

And yet, the carbon was back.

It is tempting to treat this as a failure. A flaw in the system. Bad luck.

But it is not only that.

It reveals something about how the system works in the first place.

Carbon does not disappear once it has been emitted. Most of the time, it is not stopped at source. It is handled. Stored, shifted, deferred. Moved between places, and often between people. Held in forests, in rock, in soil and in futures that have not yet arrived.

Sometimes those arrangements hold. Sometimes they don’t.

Where does it go

We tend to imagine carbon in a very simple way.

We emit it, and then we reduce it. The line on a graph rises and then — through policy, technology and collective effort — it begins to fall. This is the shape that sits behind net zero, a promise that emissions will be brought down to a point where whatever remains can be balanced out.

It is a relatively recent idea. It emerges from climate modelling and carbon budgets — ways of calculating how much carbon can be added to the atmosphere before certain temperature thresholds are crossed. Stay within the budget and warming stabilises. Go beyond it and the consequences become harder to predict, harder to live with.

But in practice, many pathways do not simply stay within that limit. They pass it.

Emissions continue in the present with the expectation that they will be drawn back down later. This is sometimes described as “overshoot” — a temporary exceeding of the limit followed by a return. Carbon is allowed to accumulate now on the understanding that it will be removed in the future.

The line still bends. Just not yet.

But that picture leaves something out.

Most of the time, carbon is not stopped at the point where it is made. It is reorganised after the fact.

Captured and pushed underground. Balanced against a forest somewhere else. Folded into supply chains that stretch far beyond the point of use. Written into models that assume it will be dealt with later.

It moves between places. Between systems. Between present and future.

It also moves between people. Some are asked to store it, some to absorb it and some to live alongside it, while others continue to produce it on the condition that it will be handled elsewhere.

A community near a proposed carbon storage site may live alongside pipelines and industrial infrastructure for decades. Meanwhile, a corporation elsewhere reports lower emissions on paper.

Carbon changes form. It changes location. It changes who is expected to take care of it.

But it does not go away.

Carbon afterlives

This is what I mean by carbon afterlives.

Not a metaphore, but a way of paying attention.

Carbon continues. It persists in different states — sometimes held in place, sometimes moving, sometimes leaking out of the systems meant to contain it. It becomes something that has to be managed over time.

That management is uneven. Some places are made to store it. Some are asked to absorb it. Some are allowed to keep producing it on the understanding that it will be handled elsewhere.

Carbon moves. Responsibility moves differently.

When it shows itself

Most of this is easy not to see.

At the point of use, things appear to improve. A building reports lower emissions. A flight is labelled “carbon neutral.” An electric car produces no exhaust pipe emissions. A supermarket promotes plant-based alternatives as climate friendly.

The numbers begin to move in the right direction.

What those numbers rely upon often sits elsewhere, lithium extraction, shipping systems, carbon offset forests, industrial agriculture, pipelines carrying compressed CO₂ beneath the North Sea.

Carbon moves. The appearance of cleanliness often depends upon infrastructures becoming less visible.

These arrangements stretch across distance and duration. They depend on systems continuing to hold.

Most of the time, they remain intact on paper.

You tend to notice them when they don’t.

A forest that has already been counted burns. Carbon stored over decades returns to the atmosphere in days. A credit issued on the basis of future growth no longer corresponds to what is actually there.

The accounting remains. The carbon does not.

In that moment, something shifts. Not dramatically, but enough.

The idea that carbon has been dealt with becomes harder to maintain.

What follows

Over the next eight weeks, I want to stay with these moments and the systems around them.

Not to arrive at a single answer, but to follow what happens after carbon is produced. Where it is sent. How it is kept in place. What happens when it isn’t.

Because once you begin asking where carbon goes, it becomes difficult to believe that it simply disappears.

We tend to tell ourselves a reassuring story about climate change. It goes something like this, emissions rise, we realise the danger and then—through policy, innovation and collective will—we bring them down. A line bends. A correction is made.

But this is not how things are unfolding.

Climate action does not proceed neatly along a line. It looks far more like a palimpsest—a surface written over again and again, where older inscriptions are never fully erased. New strategies are layered on top of existing systems, which continue to shape what is possible.

This is where the idea of mitigation deterrence—or climate delay—becomes useful. Not as an exception or failure, but as something built into the way climate governance now works.

The problem with flexibility

Nearly 90% of global emissions are now covered by net-zero pledges (Markusson et al. 2024). These pledges are made primarily by national governments, alongside cities and large corporations—not individuals. So when we talk about “net zero”, we are really talking about how countries and major economic actors say they will balance their emissions in the coming decades.

At face value, this looks like progress: a shared commitment to act.

But these pledges vary enormously. Some focus on cutting emissions directly—using less fossil fuel, changing infrastructure, reducing production. Others rely heavily on something called carbon removal.

Carbon removal refers to a set of methods for taking carbon dioxide out of the atmosphere and storing it elsewhere. This might involve planting trees, restoring soils, or using new technologies that capture carbon from the air and store it underground. In theory, this allows emissions that continue today to be “cancelled out” later.

And it is here that things become slippery.

Carbon removal introduces what policymakers often describe as flexibility. Instead of reducing all emissions immediately, countries can plan to balance some of them out in the future through removal.

But flexibility, in practice, often becomes ambiguity. It allows emissions cuts today to be weighed against hypothetical removals tomorrow. It stretches commitments across time and space. It makes it possible to promise more while doing less in the present.

The issue is not only that carbon removal technologies are uncertain—many are still unproven at scale, and some require vast amounts of land, energy or resources. It is that the expectation of their future success already shapes decisions today.

In other words, we do not just plan for carbon removal. We begin to act as if it is already available.

Plastics and the circular carbon story

A clear example can be found in the petrochemical industry—the sector responsible for producing plastics, fertilisers and many other carbon-based materials.

Traditionally, this industry has relied on fossil fuels (oil and gas) as its raw material. So if we take climate targets seriously, one obvious conclusion would be that production needs to shrink.

But that is not the direction being emphasised.

Instead, the industry increasingly frames its future through what has been called a “circular carbon” story. The basic idea is simple, carbon itself is not the problem—where it comes from and where it ends up is. If carbon can be captured, reused and kept in circulation, then production can, in theory, continue.

In this vision, plastic production does not need to fall. It simply needs to change its inputs.

Rather than extracting carbon from fossil fuels, companies suggest that they could one day use carbon captured from the atmosphere—through technologies like Direct Air Capture—or recycled from existing materials. Plastics can then be described as compatible with climate targets, not because less is produced, but because the carbon is imagined to be part of a closed loop.

It is an appealing idea. It promises continuity without sacrifice. But it rests on several quiet assumptions:

• that large-scale carbon removal will become technically and economically viable

• that storing or reusing this carbon will be secure and long-lasting

• that the environmental and social impacts (land use, energy demand, water) will be manageable

• that reducing production itself is either unnecessary or politically unrealistic

• and, crucially, that the costs and burdens of this system can be unevenly distributed—that land for tree planting or bioenergy can be found elsewhere, that energy and water demands will not fall on already strained regions, and that communities asked to host these projects will have little say in the process

This last point is often left implicit. Many carbon removal strategies rely on large areas of land, frequently in the Global South, or in rural and marginalised regions. The benefits—continued production, continued consumption—tend to accrue elsewhere. What looks like a technical solution is also a question of who bears the risk, and who gets to continue as before.

At the same time, alternative approaches—particularly those involving less production and consumption—receive far less attention. They are often framed as disruptive, costly, or simply outside the bounds of what is considered feasible.

What results is a kind of reframing: the problem is no longer how to produce less, but how to produce differently at some point in the future.

For this logic to hold, the future has to be made to feel solid—measurable, governable, already half-built.

This is where all the quiet technical work comes in – measurement, reporting and verification. Carbon removal is turned into numbers, into projections, into scenarios that can sit neatly inside policy documents and corporate strategies. Even when the technologies themselves are not yet working at scale, their anticipated effects are folded into plans as if they were.

In this way, future removals begin to count in the present. They are budgeted, traded, balanced against ongoing emissions.

But this way of knowing is not neutral. It privileges what can be quantified and modelled, what can be priced and exchanged. At the same time, it pushes uncertainty to the edges—questions about whether these systems will work, at what cost and for whom. It gives a sense of precision to something that is still, in many respects, speculative.

Carbon removal is not just a set of technologies, then. It is also a way of organising knowledge—one that makes it easier to delay, because the delay itself appears calculated, accounted for.

And this connects to something broader about growth.

We often think of growth under capitalism as simple expansion: more energy, more production, more consumption. But what is happening here feels different. It is less about straightforward increase, and more about layering.

New climate strategies—net zero targets, carbon removal, circularity—are not replacing existing fossil systems. They are being laid over them. Oil companies invest in carbon capture. Agricultural systems are reframed as carbon sinks. Plastics become part of a circular economy.

The older structures do not disappear. They persist underneath, shaping what the new layers can do.

This is why reduction feels so difficult to grasp in practice. It is not simply resisted; it is continually deferred, absorbed into new frameworks that promise change without requiring a break.

We tend to talk about this in terms of carbon lock-in—pipelines, power stations, long-lived infrastructure. And those are real constraints. But there is another kind of lock-in that is less visible.

Economic interests fold carbon removal into existing business models, allowing companies to extend rather than wind down their operations. Policy frameworks blur the line between reducing emissions and removing them later, treating them as interchangeable. And slowly, certain assumptions settle in without much challenge.

That some emissions are simply unavoidable.
That growth must continue.
That technology will compensate for what we cannot—or will not—do now.

These are not just technical positions. They are ways of seeing the world, shared across institutions and sectors. They shape what feels realistic, what feels politically possible, what feels too disruptive to contemplate.

And they are remarkably hard to shift.

There is, perhaps, a parallel here with older forms of power. Colonialism, for instance, was not only a system imposed from the outside; it also reshaped how people understood the world and their place within it. In a different way, climate lock-in operates both materially and mentally.

We inherit a worldview in which extraction is normal, growth is necessary, and substitution is preferable to reduction. Carbon removal sits comfortably within this. It offers continuity. It reassures us that the system can adjust without fundamentally changing.

So mitigation deterrence is not only about delay in a narrow sense. It is about the persistence of a particular imagination—one in which the future is always available to compensate for the present.

There is no neat conclusion to this.

These dynamics are not the product of a single bad decision or a single actor acting in bad faith. They emerge from overlapping interests, infrastructures and ways of knowing.

But if we want to understand why emissions are not falling as quickly as they should, we have to look beyond targets and promises. We have to look at the frameworks that make delay feel reasonable—sometimes even responsible.

And perhaps the question that follows is not only how to reduce emissions, but how to loosen the grip of these imaginaries.

Because if they can hold us in place, they might also be where something begins to shift.

References

Buck, H.J. et al. (2023) ‘Countries’ long-term climate strategies fail to define residual emissions’, Nature Climate Change, 13(4), pp. 317–319. Available at: https://doi.org/10.1038/s41558-023-01614-7.

Carton, W. (2019) ‘“Fixing” Climate Change by Mortgaging the Future: Negative Emissions, Spatiotemporal Fixes, and the Political Economy of Delay’, Antipode, 51(3), pp. 750–769. Available at: https://doi.org/10.1111/anti.12532.

Carton, W. et al. (2023) ‘Is carbon removal delaying emission reductions?’, WIREs Climate Change, 14(4), p. e826. Available at: https://doi.org/10.1002/wcc.826.

Lund, J.F. et al. (2023) ‘Net zero and the unexplored politics of residual emissions’, Energy Research & Social Science, 98, p. 103035. Available at: https://doi.org/10.1016/j.erss.2023.103035.

Markusson, N. et al. (2024) ‘Carbon removal and the empirics of climate delay’, Environmental Science & Policy, 161, p. 103884. Available at: https://doi.org/10.1016/j.envsci.2024.103884.

In recent years, a subtle shift has taken place in climate discourse.

For a long time, climate change was framed as something that could still be avoided—if action was taken quickly enough. The goal was prevention: limit warming, stay within thresholds, stabilise the system.

That framing is becoming harder to sustain.

Scientific models increasingly suggest that exceeding 1.5°C—at least temporarily—is likely. In response, a new set of ideas has emerged to make sense of this situation: overshoot, derailment risk, and more recently, “climate realism.”

At first glance, these appear to be technical terms. But they are also ways of governing the future.

“Overshoot” reframes the failure to stay below 1.5°C as something temporary and manageable. Temperatures may rise above the threshold, but can later be brought back down through rapid emissions cuts and large-scale carbon removal. Failure becomes delay.

“Derailment risk,” by contrast, introduces a more fragile picture. It suggests that exceeding thresholds may trigger cascading disruptions—economic, political, institutional—that undermine our capacity to respond. The issue is no longer just the climate system, but the stability of governance itself.

Both framings attempt to deal with the same underlying problem: how to maintain political commitment in the face of likely failure.

This is where “climate realism” enters.

Rather than focusing on prevention, climate realism shifts attention toward managing a world of ongoing disruption—one defined by instability, adaptation, and strategic risk. The emphasis moves from avoiding crisis to living within it.

Taken together, these ideas point toward a deeper transformation.

Climate change is no longer governed primarily as a problem to be solved, but as a condition to be managed.

In this context, something like a new version of “There Is No Alternative” begins to emerge.

Not the neoliberal TINA of market inevitability—but a quieter, more technical version:

the sense that overshoot, disruption, and constrained responses are simply unavoidable

This is what might be called Eco-TINA.

Its power lies in how it operates. It does not deny the severity of the crisis. In fact, it intensifies it. But in doing so, it can narrow the space of political imagination.

If overshoot is inevitable, if disruption is unavoidable, if governance must focus on managing risk rather than transforming systems—then the range of possible responses begins to contract.

Technocratic interventions become more acceptable. Large-scale carbon removal, geoengineering, and security-led approaches are justified not because they are ideal, but because alternatives are framed as unrealistic.

The danger here is not that limits are real—they clearly are.

It is that the language of limits can begin to do political work. It can make certain futures seem inevitable, and others unthinkable.

The question, then, is not whether climate change involves risk, uncertainty, or constraint.

It is:

how those conditions are interpreted—and what they allow us to imagine, or prevent us from imagining, in response.

Subscribe now

From net zero targets to carbon capture technologies, many proposed responses to climate change appear rigorous and actionable. But look more closely, and they often rely on assumptions, deferrals, or accounting practices that seem to reproduce the very conditions they are meant to transform.

This isn’t simply a failure of ambition or political will. It points to something more structural.

When a problem like climate change threatens foundational systems—economic growth, state sovereignty, industrial infrastructure—responses are not generated outside those systems, but within them. As a result, solutions are shaped to fit existing logics, even as they attempt to challenge them.

You can see this pattern in several places.

Many climate mitigation pathways depend on large-scale carbon removal technologies that do not yet exist at the necessary scale. This allows emissions to continue in the present, on the assumption they can be reversed in the future. Responsibility is displaced forward in time.

At the same time, emissions are counted territorially—within national borders—even though production and consumption are globally distributed. This can create the appearance of reduction in one place while emissions are displaced elsewhere.

And technologies such as carbon capture do not simply remove carbon. They reorganise it—moving it through pipelines, storing it underground, embedding it within infrastructures that extend its life rather than resolving it.

Across these examples, a similar dynamic emerges.

Climate action is translated into forms that existing systems can process. The result is a series of persistent tensions:

• between present action and future deferral

• between national targets and global realities

• between decarbonisation goals and fossil-fuelled infrastructures

These tensions are not hidden. But nor are they fully addressed. Instead, they recur across policy, modelling and public discourse as partially visible, often incompatible realities.

At the same time, social and political conditions make it harder to engage with this complexity. Climate change is not characterised by a lack of knowledge, but by a fragmentation in how it is understood and acted upon.

So the question shifts.

Not simply: Are we solving climate change?
But:

What kinds of solutions are we capable of producing within our current systems—and what gets left out, deferred or reshaped in the process?

This space is for working through that question.

Subscribe now