The conversation with Žiga Malek was initiated on the basis of the 2017 PNAS article Natural Climate Solutions (NCS), which he suggested as a starting point for discussing the quantified mitigation potential (and limitations) of land-based climate action. The article examines how much NCS, such as conservation, restoration and improved land management, can contribute to reaching the Paris Agreement goals, to hold the global temperature rise below 2 °C. Authors show that NCS can provide over “one-third of the cost-effective climate mitigation” that would be needed to be taken between 2017 and 2030 to stabilise warming, while providing co-benefits, such as improving soil productivity, maintaining biodiversity, and the quality of air and water. The rest of the climate mitigation strategy should be provided by “aggressive fossil fuel emissions reductions”.

Žiga Malek is a land systems scientist who is building models of potential changes in land use in order to inform the clients and policy makers, and to steer towards a more sustainable land use in future. He consulted organisations such as the European Environment Agency, the World Bank, and Fairtrade International, among others. In the following conversation, we speak with Malek about the potential of NCS.

When we ask how much climate mitigation “nature” can provide, what is meant by nature in this context? What are “natural climate solutions” as opposed to “nature-based solutions”?

It depends on who you ask. From a land-cover perspective, nature might be defined as areas not directly altered by human activity. But roughly 70 percent of the ice-free terrestrial surface has already been modified. Even ecosystems that appear untouched are indirectly affected by climate change, pollution, or historical land use.

We did a study a few years ago with my colleagues Katerina Schultz and Peter Verburg and mapped how many areas we consider “natural” have been altered historically, for example by burning, grazing, or agriculture centuries ago, later abandoned, and then reforested. So “nature” is relative. Compared to a city, it is natural. Compared to a primary forest, it is heavily modified.

As for terminology: “Nature-based solutions” is a broad category. It includes interventions that work with ecosystems rather than mechanical carbon removal technologies, such as Direct Air Capture (DAC). “Natural climate solutions” are more specific. They are nature-based actions explicitly targeted at climate mitigation: forest conservation, reforestation, wetland restoration, improved grazing, nutrient management in croplands. Nature-based solutions can also be used for many other objectives, like cooling cities, managing stormwater runoff, or improving urban liveability. Natural climate solutions are framed primarily around climate, often at much larger scales.

And importantly, NCS is not only about planting trees as a “natural measure”. In agriculture, we can reduce emissions and increase soil carbon through different practices: improved fertilisation patterns, reduced tillage, changing crops, and improving grazing management. Overgrazing is a major issue in many places: too many animals per hectare can degrade ecosystems; then pressure shifts elsewhere, creating chain reactions. Better management can be a climate solution, even though it is not “natural” per se.

It is surprising, however, that NCS can contribute over one-third of the climate mitigation.

If we talk about large contributions like “30% by 2030”, that implies very large-scale global interventions. Many of us are sceptical that we can achieve this while maintaining current lifestyles.

We tested this in China: what if China wanted to store carbon through ecosystems, forests and restored wetlands, while maintaining high demand for meat and dairy, space for cars and cities, and so on. It would not work. These measures require millions of hectares, and we don’t have such space. Land is finite. If you allocate land to forests and wetlands to offset emissions from transport, something has to change in cropland and grazing. There is a conflict between ever-increasing land demand and the idea that we can “plant trees” to solve climate issues, without changing anything else. Ecosystems are useful, but we cannot rely on them as a single solution.

Often, the regions with the highest carbon-sequestration potential are also where land demand is highest for soy expansion, mining for electronic components, infrastructure, and so on.

So the reforestation and “hype” around planting trees, as forests being the most cost-effective NCS, is not as viable as it seems.

What the authors of the mentioned study did, and what we often do, is to look at the actual potential of NCS. We often hear politicians say, “We will plant trees”, or “we will create wetlands ”, or restore some other degraded habitat. But there are examples where planting trees can have negative effects: in South Africa, large pine plantations reduced downstream water availability. In Portugal and Spain, eucalyptus can sequester carbon quickly, but it also changes hydrological cycles and fire regimes, and increased fire can release more carbon emissions. So it is not as simple as “let’s plant trees”.

“The largest potential of NCS lies in avoided deforestation and natural forest management, not in reforestation. But that doesn’t bring political points, it is not sexy, while prevented deforestation is truly a heroic act.”

We plant fast-growing exotics like eucalyptus, while simultaneously cutting down native forests that store far more carbon, such as the Amazon or the Cerrado. That is incoherent. The paper, Natural Climate Solutions, emphasises that the largest potential lies in avoided deforestation and natural forest management, not in reforestation. But that is not photogenic or sexy. Investors and politicians take pictures planting a tree, not standing in front of an intact forest saying, “We prevented deforestation.” Yet that is the truly difficult, heroic act. An old-growth forest left intact can outperform repeated short-rotation plantations. Old-growth systems store more carbon, maintain higher biodiversity, and require less fertiliser and irrigation. And every harvest cycle disturbs soils, releasing carbon.

“We are entering a negative feedback loop: fossil fuel use and ecosystem degradation warm the planet, which then undermines ecosystem function, which then reduces the capacity to rely on ecosystems to store our carbon.”

Since that paper came out in 2017, we have also seen that some forests we assumed were carbon sinks are becoming carbon sources — parts of the Amazon, Congo, Indonesia, and even Finland — due to forest modification, conversion to agriculture, altered species composition, drought, fires, and climate stress. That does not mean “cut them down”; it means a warning we are entering a negative feedback loop: fossil fuel use and ecosystem degradation warm the planet, which then undermines ecosystem function, which then reduces the capacity to rely on ecosystems to do what we want them to, to store our carbon.

That makes it even harder to predict outcomes. Also, trees in areas that worked in the past may not work under new climate conditions.

Exactly. It means our efforts in other sectors must be more forceful, especially in reducing fossil fuel use. Natural climate solutions are necessary, but they are not a substitute for decarbonising energy, transport, and industry.

We have to be strategic. These solutions need to persist for decades. If you plant trees and the next government cuts them down, or if trees die because they were planted in the wrong place without care, the climate benefit vanishes. There was a famous campaign in Turkey, where they planted a record number, 11 million trees, and around 97% died within the first year because of poor conditions and lack of care.

This is why there was backlash against “planting trees everywhere”. There is a paper titled The Tyranny of Trees, and some politicians reacted as if ecologists “hate trees”. But the critique is not anti-tree; it is anti-misplaced tree planting. Planting trees in the wrong ecosystems can cause harm. For example, proposals to plant trees in savannas, which are natural habitats for elephants, rhinos, cheetahs, and lions, can destroy those ecosystems by altering hydrology and drought dynamics. Not everything valuable is a forest.

Wetlands, on the other hand, can be more carbon-dense than forests. How come?

Simply put, wetlands often have wet, oxygen-limited soils, and carbon accumulates over long periods. Even treeless wetlands can store massive amounts of carbon in deep soils. If you drain them, much of that stored carbon is released to the atmosphere, which is a major problem. Wetlands cover a small fraction of land yet hold a disproportionate share of soil carbon. This carbon density is why protecting and re-wetting wetlands matters. But we cannot simply turn “all of Europe into wetlands”. We have already lost more than half of Europe’s wetlands (and globally too), and restoring them is difficult because those areas are now cities, infrastructure, or agricultural land.

Wetlands also suffered historically from negative perceptions. They were “swamps”, with mosquitoes and leeches. Using “wetland” as a term was, in a sense, a rebranding. Many ecosystems contribute much more than we assume.

A few short questions on technical aspects: Where does the data you use for climate modelling come from?

The data is synthesised from many other studies. For example, with forests, scientists measure carbon stocks using a combination of satellite data and field observations.

Satellites are one part: they show land-cover type. But they do not show species composition very well, or what is happening below ground. Depending on the ecosystem, it might be that one-third is in biomass above ground and two-thirds below ground, so scientists combine satellite imagery with field measurements. They measure carbon stored in soils across ecosystems — forest soils, agricultural soils, urban soils, wetlands, grasslands — in different regions. To allocate NCS effectively, we need to understand which contexts can sequester carbon effectively, and where they cannot, or where interventions could even do harm.

Results are typically expressed as averages or medians with uncertainty ranges. If a study estimates that a measure could store 80 million tonnes of CO₂, the range might be 40 to 120 million tonnes, depending on spatial context and implementation success.

And success is not guaranteed. Sometimes benefits only accrue over long periods, 50 to 100 years, while some systems use short rotations: fast-growing species such as eucalyptus or bamboo harvested after 10 years and replanted. That changes the accounting.

What tools do you use?

There are many models, and different research groups use different ones. Some banks and ministries have their own. They are geospatial, so you need spatial data: land-cover maps, soils, climate, and so on. The model choice depends on the client and the question. In Amsterdam we developed a model called CLUMondo, which is widely used, but there are others. Often we combine models — some operate coarsely at national scales, and others allocate changes locally.

Who is it used by?

These models are typically developed for larger spatial scales, countries, regions, continents, or global strategies, rather than municipalities. They require extensive datasets, interdisciplinary collaboration, and time, so they are relatively costly. The clients tend to be international institutions, development banks, national governments, or multilateral organisations.

For example, we worked with the Inter-American Development Bank to explore future land-use scenarios in several South American countries. The objective was a future without deforestation. The client provided demographic and agricultural production projections. We modelled alternative land-use futures — identifying where agricultural expansion might occur, which areas could be protected, and how different policy choices would alter outcomes. The client then evaluated associated costs and benefits for people, including climate regulation, soil conservation, reduced fire risk, and broader economic impacts.

It is important to emphasise that these models do not predict the future. They construct scenarios, and we often test different scenarios, comparing trajectories under different policy assumptions — for example, business-as-usual versus alternative strategies, and evaluate synergies and trade-offs for climate, biodiversity, water, food production, and economic systems.

Then there is time commitment. The modelling horizon is usually mid-term, often to 2050. However, unexpected geopolitical or economic events — such as wars or trade disruptions — can rapidly alter land demand and introduce new uncertainties.

Some practical examples or effects of incorporating such models, and why measures are not taken more aggressively?

Loans can be conditioned by environmental performance: the funding granted on the basis that agricultural practices would be improved, forests better protected, certain areas set aside, and broader gains delivered for climate, biodiversity, and water systems, as indicated by the modelling results.

Many examples are long-term, so we will only know outcomes properly in five to ten years. Often, modelling is a first step; the next step is to downscale to local decisions and policy, implementation details.

What are you currently working on?

I am working on land-based mitigation and natural climate solutions in Europe. I collected hundreds of case studies on NCS and examined whether they improved carbon and soil characteristics, biodiversity, or farmers’ income, or if they caused negative trade-offs.

I also investigate the spatial context: what type of soils, climate, where benefits are likely, where risks are likely — and I hope this will be useful for other scientists and policymakers. For example, if billions of euros are allocated for farmers to plant trees, we need to know where this makes sense, as it may fail in many places. And national contexts differ: Slovenia already has about 61% forest cover; the Netherlands around 6%. The opportunity space is not the same. The solutions need to be tailored — including species choice and management. So the goal is more targeted, context-specific measures.

We can be disappointed after the COP 30, because fossil fuel phase-out plans were not agreed upon. And also in the European Union, we see contradictions. The regulations to prevent importing products linked to deforestation are at odds with trade dynamics.

We import a lot, not only beef, but feed crops, coffee, cocoa, and so on, and these supply chains can be highly damaging.

If eight billion people had a European lifestyle, it would be a disaster, even if we want welfare and good lives for everyone. It also matters that Europe’s and the US’s share of current annual emissions is lower than their historical contribution. The CO₂ from past emissions is still in the atmosphere. Then there are embedded emissions, exporting emissions through importing goods from China, for example.

We should be more ambitious and stop treating climate policy as “damaging our economy”, when our economy has already damaged the global carbon cycle and biodiversity. China, India, Nigeria and others are adopting renewables earlier in their industrial trajectory than Europe did, 150 years after the industrial revolution. They are not moving fast enough, but given that they are not starting from the same historical position, they are doing a better job than Europe.

I need to ask this: isn’t it interesting that European dependency on imports of Russian gas gave the incentive to reopen coal mines? How could we collectively fail to the point of sending people back to work in mines instead of, for example, deciding on the common benefit of heating less?

I think studies showed that such drastic measures would not even have been necessary. If we reduced heating by 10% and used public transport more, we could have managed the crisis. Since we imported a large share of fertilisers and livestock feed from Ukraine and Russia, the European Parliament also permitted the use of previously banned glyphosate and other fertilisers. Yet studies indicated that if we consumed 10% less meat for a few years, we would not have needed these measures.

Ten percent is not a lot. I would not consider it a sacrifice. It is logical and necessary. If we are not able to make such decisions in times of crisis, it becomes even less likely that we will make them in our everyday lives.

“By reducing energy and food consumption by 10 %, we could avoid a crisis. It is not a sacrifice, it is necessary, fair and logical”

If you had a chance to rule the world, what rules would you enforce first?

I would start with low-hanging fruit: living healthier and reducing excess. Not banning bananas or chocolate, not at all, but evidence suggests that if Europeans followed national health guidelines (still including meat, dairy, even some alcohol), we would save an area of cropland roughly the size of France. That means lower emissions, less fertiliser, less fossil fuel use.

So we can live virtually the same, but a bit healthier and less excessive, and save a lot. That frees capacity to focus on the difficult tasks: where to restore wetlands, where to plant trees to maximise carbon and biodiversity, and so on. And it also connects to justice: we compete for land globally. We buy feed from Brazil and Argentina; that land pressure has consequences. Meanwhile, hundreds of millions of people are hungry.

After COVID, we returned to the same old patterns. Yet it was the best chance ever: the world paused.

COVID was a missed chance, but also an evidentiary breakthrough. Suddenly, we could measure, empirically, what pollution reductions looked like when traffic and some industrial activity fell. Satellites showed drops in emissions and pollutants. We could attribute changes more clearly than in normal times. Decision-makers could have said: “Now we have proof; now we act.” But COVID ended, and the lesson was largely ignored. We returned to the familiar “it’s complicated” discourse.