A 2025 paper in Nature Communications makes the strongest quantified case yet for building with mass timber at scale. But it also highlights the potential for deforestation.
For about a decade, cross-laminated timber has been positioned as one of the defining answers to decarbonising construction. The pitch is straightforward and, on its face, persuasive: engineered wood carries the structural loads that once required concrete and steel, wood stores carbon rather than emitting it, forests grow back, and a building sector that switched to timber at scale could move from net emitter toward something closer to a carbon sink. The towers exist to make the argument visible – Dalston Works in London, Mjøstårnet in Norway, Ascent in Milwaukee – and each year the tallest timber building in the world climbs a little higher, and the claim grows a little louder.
What has been largely missing from the conversation is the question of what happens when the switch is made not in a handful of showcase towers but globally, across the next seventy-five years – and that is the question a paper published in Nature Communications in May 2025 sets out to answer. Global Land and Carbon Consequences of Mass Timber Products, led by Kai Lan with co-authors from Yale, the USDA Forest Service and other institutions, does the speculative calculations for what could happen if CLT at scale is achieved.
Most mass timber lifecycle studies examine a single building or a single supply chain and ask whether one timber tower is lower-carbon than its concrete equivalent, and usually it is. This is a useful answer to help sell the benefits of CLT and shift behaviour in a stubborn construction sector, but it’s the wrong tool for a systemic question.
The methodology combines two instruments: a life cycle assessment, which traces the material’s carbon from forest to end-of-life, and the Global Timber Model, an economic model of how the world timber market responds when demand shifts.
Most assessments assume a supply of sustainable wood sitting ready to be drawn on; the Global Timber Model is based on supply and demand. It models what the market actually does as prices move – which forests get planted, which get harvested, which get left standing – the way managed supply governs the price of any traded commodity. That difference turns out to matter a great deal.
The paper models three rates of CLT adoption – low, medium and high – from 2020 to 2100, against a baseline in which mass timber never really takes off. In the high scenario, where CLT replaces concrete or steel in roughly 60 per cent of new urban mid- and high-rise buildings, the numbers are large. Cumulative lifecycle greenhouse gas reductions reach up to 39 gigatonnes of CO2 equivalent by the end of the century – roughly a year of current global energy-related emissions, avoided or stored by a change of material.
Productive forest area expands by up to 36.5 million hectares, close to the land area of Germany, and long-term carbon storage rises by 20 to 25 gigatonnes, held partly in the buildings themselves and partly in the forests grown to supply them.
These are the numbers from the abstract that point to nothing but a positive outcome. However, this isn’t where the story ends.
The forest expansion the model predicts doesn’t happen evenly. It concentrates in countries that already have established plantation forestry and mature timber industries – the United States, western Europe, China, Canada – where rising prices pull more land into managed forest, shorten rotations and favour faster-growing species for higher yield per hectare.
In the tropics, the same price pressure runs the other way. Near the equator – the Amazon, Borneo, the forests of Ecuador – where forestry controls are weaker and land is cheaper, the model predicts natural forest decline. The authors put it plainly: there is, they write, “a tradeoff between conservation and carbon goals.”
In places with cheaper land and lighter regulation, the timber for all these new panels comes not from new plantations that replace existing natural forests.
Which means looking closer at the word forest. A monoculture pine plantation stores carbon, and the paper is right to count it; by every metric the assessment uses, it contributes. But a plantation is not a forest in the sense most people mean. It does not shelter the same biodiversity, hold the same soil ecology, regulate water the same way, or carry the history of a place. It is, more accurately, an agricultural crop that happens to be made of trees.
This is what makes the paper more useful than a verdict would be: mass timber at scale is a policy problem. The carbon benefit is real and it is conditional, and the condition is provenance. A timber panel is only as sustainable as the supply chain behind it, and nobody building a tower buys a forest. They buy panels from a manufacturer, who buys lamstock from a mill, who buys logs from a concession holder who may or may not be cutting from the forest the specifier imagines.
Certification and provenance, in this chain, do far more than paperwork; they carry the whole of the carbon claim. There are plenty of locally managed plantations that leave standing forest alone and return its profits to the community, but it does mean the right question to ask of a piece of mass timber is not whether it is mass timber – but rather, where did it come from?
The paper is candid about its limitations, and they’re worth keeping in mind. Its economic model assumes functioning markets and reasonable enforcement; in regions where illegal logging is already entrenched, that assumption is generous. It models carbon and land area, not biodiversity loss, water impact or soil health – a plantation and a natural forest can hold the same tonnage of carbon and be nothing like the same ecosystem. It does not account for the adhesives that bind the panels, whose formaldehyde emissions, petrochemical dependence and end-of-life complications are a story of their own. And it runs on the assumption that the next seventy-five years behave roughly like the last few decades, that bushfire, climate-driven dieback, pest outbreaks and policy reversals stay within historical bounds – an assumption that loosens every year. None of this invalidates the work. It makes it a starting point rather than a conclusion.
There’s no perfectly sustainable material, only the right compromise for the right situation – and mass timber can be a very good one, in the right building, with the right supply chain, held to the right standard.
