Designing cities to cut transport infrastructure embodied emissions

The planet’s transport systems are on the cusp of one of the largest construction surges ever seen.

As populations rise and economies expand, countries will need enormous quantities of new roads, railways, bridges and airports.

If that demand is met in the usual way, it could require hundreds of billions of tonnes of concrete, steel and asphalt by 2060.

A new study concludes that the biggest chance to cut emissions is not primarily about making construction cleaner. Nor is it simply about moving everything on to rail.

Instead, the largest climate benefits come from planning cities so that people require less infrastructure in the first place.

A world built on roads

The analysis was led by André Baumgart, a doctoral researcher at the University of Natural Resources and Life Sciences, Vienna (BOKU).

Using crowdsourced information, his team charted global road and rail networks at a resolution detailed enough to identify individual streets.

They then linked every segment to the quantities of concrete, steel, asphalt and gravel contained within it. For 2024, the total stock amounted to roughly 355 billion tonnes of material.

Road infrastructure accounts for the vast majority. Around half of the mass sits in lower-grade local and rural roads, while higher-grade roads make up most of the remainder. Sand and gravel by themselves represent almost three-quarters of the total.

Rail is a much smaller share-about 6 per cent of the material stock. Even so, it contains more than two-thirds of the steel, because tracks and stations rely heavily on it.

These stocks are distributed very unevenly across the world. In densely populated, lower-income regions such as Sub-Saharan Africa and South Asia, they remain under 5 tonnes per person.

By contrast, in more spacious and wealthier regions such as North America, they can rise above 500 tonnes per person.

Infrastructure is about to surge

By 2060, demand for mobility is expected to increase by around 80 per cent, broadly following trends in population and income.

If the world satisfies that demand using today’s dominant approach, the material stock in transport infrastructure would roughly double to about 720 billion tonnes.

Reducing the disparity between richer and poorer regions would raise totals further. If all countries were to reach a reasonable minimum level of access, global stocks would triple to more than a trillion tonnes.

Most of that expansion would take place in the Global South-lower-income regions that are urbanising rapidly. By 2060, South Asia’s material stock would rise eightfold, while Sub-Saharan Africa’s would increase sevenfold, as both build to close the gap.

This projection also prompts a basic question: how much infrastructure is actually necessary?

One study covering well over a hundred countries found that the benefits of mobility stop improving once a moderate level of material stock is reached. Beyond that point, adding more roads contributes little to people’s lives.

On the reference trajectory, the world would also seal roughly 518,000 km² of additional land, fragmenting habitats as it goes. Growth at this scale is not merely theoretical.

Over three decades, China more than tripled its road and rail stocks-evidence of how quickly infrastructure can expand when an economy accelerates.

Rail’s hidden carbon cost

A seemingly straightforward solution is to build railways instead of roads. Rail can move more people using less land, and per trip it typically consumes far less fuel. However, the study highlights an easily overlooked downside.

Track construction is extremely material-intensive. Rails, sleepers, bridges and stations demand substantial volumes of concrete and steel. Producing those materials releases carbon well before any trains begin operating.

This up-front climate impact is known as embodied emissions-the carbon associated with making and building something, rather than running it. In the study’s scenarios, diverting future road expansion towards rail would more than triple embodied emissions by 2060.

In their modelling, embodied emissions rise to about 37 billion tonnes of CO₂, compared with 11 billion tonnes under the current pathway. Nearly all of the increase comes from concrete and steel.

This does not mean rail is the wrong option. Trains can still reduce emissions from everyday travel, and over time those operational savings may outweigh the extra concrete.

The key point is that an apparently cleaner transport system can carry a higher carbon burden at the moment it is built-something that is easy to miss when analysis focuses only on tailpipe emissions.

The biggest lever is less travel

Because more rail can increase embodied carbon, the strongest lever lies elsewhere. The study’s clearest climate message is to cut how much infrastructure must be constructed in the first place.

That largely depends on urban form-how cities are arranged. Compact, mixed-use neighbourhoods place homes, shops and workplaces closer together, reducing trip lengths and lowering the need for extensive road networks.

Urban sprawl works in the opposite direction, demanding far more asphalt per person to provide the same level of access.

Reducing travel demand cuts embodied emissions from infrastructure by about 44 per cent by 2060, bringing them down to roughly 6 billion tonnes of CO₂. Most of that decrease comes from denser urban development.

Two widely discussed approaches have little effect in this context. Cutting the car fleet in half through sharing, or switching every vehicle to electric, changes relatively little because both mostly affect vehicles and parking-only a small fraction of the concrete and asphalt involved.

The advantages of density are not limited to climate outcomes.

“Compact, transit-oriented, and walking- and cycling-friendly cities can deliver access to mobility with far less material and carbon than car-centric sprawl,” said Baumgart.

Cleaner materials still matter

Demand-side measures are only part of the picture. The other side is how the remaining concrete, steel and asphalt are produced.

Lower-carbon manufacturing, increased recycling, substituting materials and deploying carbon capture can all reduce the emissions embedded in new construction.

By themselves, these measures deliver only modest reductions-cutting just 4 to 8 per cent from the total. Achieving bigger gains depends on a broader effort to decarbonise the heavy industries that supply construction materials.

A separate global study reached a comparable conclusion for buildings and cars, finding that recycling and efficient design avoid the most carbon when they are combined with clean energy.

The best way to cut emissions

When the two approaches-less travel and cleaner, more recycled materials-are combined, the benefits reinforce one another.

Together, they can reduce embodied emissions from future infrastructure by 63 per cent, while still ensuring that every region achieves a reasonable level of access.

The significance is clear when set against the carbon budget-the remaining CO₂ the world can emit while still limiting warming. Continuing to build as the world does today would use up a few per cent of that budget.

A strong swing towards rail performs worse on this measure: the concrete alone could lock in close to a quarter of the budget associated with the stricter 1.5°C (2.7°F) target.

The researchers frame this as a way to align aims rather than choose between them.

One of the study’s co-authors, Volker Krey, is a climate researcher at the International Institute for Applied Systems Analysis (IIASA).

“Meeting the world’s growing mobility needs and achieving climate goals can appear to be competing objectives, but integrated planning can help reconcile them,” said Krey.

Less travel beats cleaner construction

Climate models have seldom tracked how growing demand for mobility translates into physical roads and railways-and what that implies for carbon emissions.

This study sets out that link explicitly for the first time, from how far people travel, to tonnes of concrete, to tonnes of CO₂.

Its central takeaway is that the largest savings come from needing less, not merely from building in a cleaner way. Cutting travel demand outperforms any single technology substitution. The greatest reductions come from combining demand reduction with cleaner materials.

The authors also emphasise that decisions about what gets built, and where, matter more than sheer volume. Once a city’s form is established, it shapes how residents travel for generations.

Cities built for the future

That is why the opportunity is greatest now-especially in regions where cities are expanding rapidly.

“Choices made now in rapidly urbanizing regions will lock in mobility patterns for decades due to the long service lives of infrastructure,” said Baumgart.

For planners in fast-growing cities, the study reframes the challenge as an active choice.

Denser neighbourhoods, safe walking and cycling networks, and rail where it replaces car travel can provide the access people need with a fraction of the carbon impact.

What is built in this decade will still be in place in 2060, and much of its climate cost is being determined today.