The management choices behind Eucalyptus plantations

More trees or bigger trees?

Every plantation begins with a deceptively simple decision: how many trees should be allowed to grow? Plant many trees close together and the site is rapidly occupied, producing a large amount of wood per hectare. Give individual trees more space—or remove competitors through thinning—and something different happens: growth becomes concentrated in fewer stems, allowing them to become larger much faster. This trade-off lies at the heart of plantation forestry. In our study, Thinning regimes and initial spacing for Eucalyptus plantations in Brazil, we examined this question using 188 experimental plots in four clonal Eucalyptus trials in Bahia and Espírito Santo. The experiments compared different planting densities and a range of thinning strategies, from leaving stands untouched to removing most trees early in the rotation.

Before thinning, plantations established at 1,111 trees per hectare suffered considerably greater mortality than those planted at 667 trees per hectare, although early diameter growth was surprisingly similar. The major differences appeared once competition intensified. When stands were thinned early—at about 2.5 years—and only 150 trees per hectare were retained, growth became concentrated into those selected trees. By age 11.5 years, mean diameters of around 35–38 cm were achieved in these intensive regimes. The most intensively thinned treatment produced a mean tree diameter around 78% greater than the unthinned treatment. But there was a price: concentrating resources into fewer large trees reduced total stand basal area. In forestry, therefore, faster growth of individual trees does not necessarily mean greater production of wood at stand level.

There is no universally “best” thinning regime. The best strategy depends on what the forest is expected to produce. If the objective is pulp, energy or large quantities of relatively small-diameter wood, maintaining many trees can maximise stand stocking. If the objective is sawn timber or other higher-value solid wood products, early thinning can redirect the productive capacity of the site towards a much smaller number of large stems. Intermediate strategies can even produce several products during the same rotation, harvesting commercial wood in successive thinnings while keeping selected trees for a later final harvest. Plantation forestry is therefore not simply about maximising growth; it is about deciding where that growth should go. Understanding these trade-offs between biomass production, tree size, rotation length and product value remains an important part of research Biomass Production, (website).

---

Read the study

Ferraz Filho, A. C., Mola-Yudego, B., González-Olabarria, J. R., & Scolforo, J. R. S. (2018). Thinning regimes and initial spacing for Eucalyptus plantations in Brazil. Anais da Academia Brasileira de Ciências, 90(1), 255–265. https://doi.org/10.1590/0001-3765201720150453 [RG]

Full article: Read the open-access article at SciELO

What if Europe grew more grass?

A climate-soil strategy

Grass hardly looks like a technological solution to climate change. It grows beside roads, between fields and across millions of hectares of European countryside. Yet its apparent simplicity hides something interesting. Modern agriculture has increasingly specialised in annual crops, and that intensification has come with a legacy of depleted soil carbon, nutrient losses, erosion and other environmental pressures. What would happen if perennial grass were deliberately brought back into crop rotations—not simply as another crop, but as part of the environmental infrastructure of agricultural landscapes? This was the question behind our study, Large-scale deployment of grass in crop rotations as a multifunctional climate mitigation strategy. Instead of looking at a handful of experimental fields, we modelled more than 81,000 sub-watersheds across the EU27 and the United Kingdom, asking where additional grass could make sense and what it might achieve.

The interesting part is that the answer is not only about carbon. Bringing perennial grass into rotations could help rebuild soil organic carbon while simultaneously producing large quantities of biomass and protein. The same change could reduce nitrogen losses to water, limit wind and water erosion and, in some landscapes, contribute to flood mitigation. If some of the harvested grass were processed in green biorefineries, it could also become feed protein, biogas and raw material for other biobased products. In the scenarios explored in the study, the combined annual greenhouse-gas savings from soil-carbon sequestration and replacing natural gas with grass-derived biogas were estimated to be equivalent to roughly 13–48% of current greenhouse-gas emissions from European agriculture. This is not a prediction that such savings will automatically occur: it is a modelled estimate of the scale of the opportunity under widespread deployment. But it reveals an important point. A hectare of agricultural land does not necessarily have to deliver just one service. Managed differently, the same landscape can produce commodities while also storing carbon, protecting soils and improving water quality.

There is, of course, a catch—and it may be the most interesting part of the story. Growing grass is easy; creating an economic system that values everything the grass does is much harder. Farmers need markets for the biomass, biorefineries need investment and reliable supply chains, and environmental benefits such as carbon sequestration or cleaner water need mechanisms that translate them into real incentives. The study therefore points beyond a simple choice between food production and environmental protection. The challenge is to design agricultural and bioeconomy systems in which the two reinforce one another. In that sense, grass becomes more than a crop: it becomes a possible bridge between agriculture, renewable energy, soil restoration and the emerging circular bioeconomy. This connection between biomass production and the wider functions of landscapes is also central to research at the Biomass Production. Further publications and research activities are available here.

Co-benefits of introducing grass production in crop rotations with the primary objective of enhancing soil organic carbon. The figure shows the relative to contribution toward reaching the classification “low impact” at the landscape scale for nitrogen emissions to water, soil loss by water erosion, and soil loss by wind erosion, respectively, in the low-estimate (left) and high-estimate (right) scenarios. Landscapes that already have a “low” or lower impact are excluded.

---

Read the study

Englund, O., Mola-Yudego, B., Börjesson, P., Cederberg, C., Dimitriou, I., Scarlat, N., & Berndes, G. (2023). Large-scale deployment of grass in crop rotations as a multifunctional climate mitigation strategy. GCB Bioenergy, 15(2), 166–184.

Full-text article: Open-access PDF
Journal: GCB Bioenergy — Wiley
Article page: Wiley Online Library
DOI: https://doi.org/10.1111/gcbb.13015

Wind damage can trigger years of forest vulnerability

Windstorms are among the most important natural disturbances in Northern Europe, causing extensive tree mortality, timber losses and long-lasting changes in forest structure. Large-scale windstorm damage has increased across Central and Northern European forests in recent decades, while in boreal regions such as Finland, warmer winters and shorter periods of frozen soil may further reduce tree stability and increase vulnerability to wind.

The consequences of wind damage, however, can continue long after the storm has passed. A large-scale study of more than 70,000 wind-damaged forest stands across Finland found that forests affected by wind may remain vulnerable to subsequent bark beetle or snow damage in the years that follow. These disturbance pathways depend on forest structure, climate, topography and tree-species composition.

“Forest disturbances are often considered separately, but in fact, one disturbance can change the conditions under which the next one occurs,” says Doctoral Researcher Qianqian Tian, lead author of the study at the University of Eastern Finland. “We wanted to understand what happens after wind damage and whether forests subsequently affected by bark beetle or snow damage share the same characteristics.”

Forest structure played an important role in shaping what happened after wind damage. Warmer conditions and forest composition helped distinguish stands subsequently affected by bark beetles from those affected by snow, while mixed stands showed a greater tendency towards bark beetle rather than snow damage when compared with pine-dominated stands. These differences suggest that post-storm vulnerability depends not only on the initial disturbance, but also on the structural and environmental conditions of the affected forest.

“This has important practical implications, since it can help us adapt monitoring and forest management after a storm to reduce future losses,” says Professor Blas Mola-Yudego, coordinator of the study at the University of Eastern Finland. “After a storm, a dense and structurally heterogeneous forest in southern Finland may require different attention than a pine-dominated stand in North Karelia. Forest risk management should consider what may happen next, rather than focusing only on the damage that has already occurred.”

The timing of subsequent disturbances was also notable. Around 94% of wind–snow damage sequences and 76% of wind–bark beetle sequences occurred within five years of the initial wind disturbance, while more than 90% of both types of sequences occurred within eight years.

“The years immediately following a major wind disturbance can therefore be particularly important for monitoring and preventing further damage, but our results also show that we should not look for the same warning signs everywhere,” says Marina Peris-Llopis, co-author of the study.

Mari Selkimäki, lecturer in forest planning at the University of Eastern Finland, highlights the importance of the large-scale forest information used in the research: “Combining national forest data with long-term damage records allowed us to look beyond individual storms and identify broader patterns in how disturbances interact across boreal forests.”

More than 70,000 observations openly available for further research

An important outcome of the study is that the processed dataset underlying the analysis has been made openly available for reuse and further modelling. The dataset contains information for more than 70,000 wind-damaged forest stands across Finland, combining variables describing forest structure and species composition with climate, topography and subsequent bark beetle and snow damage.

The dataset can therefore be used beyond the analyses presented in the paper, for example to test alternative statistical or machine-learning approaches, investigate disturbance interactions, develop risk models, or compare post-disturbance dynamics with other forest regions.

➡️ Open dataset: https://doi.org/10.5281/zenodo.21060824

The results support more targeted monitoring after storms. Structurally heterogeneous forests containing spruce, particularly in warmer areas, may deserve greater attention for subsequent bark beetle damage, whereas forests on steeper terrain may require closer monitoring for snow damage. Maintaining stand stability and avoiding excessive stand density may also help reduce vulnerability to interacting disturbances.

➡️ Read the paper: Tian, Q., Peris-Llopis, M., Selkimäki, M., & Mola-Yudego, B. (2026). Post-wind vulnerability of boreal forests: Structural and site drivers of sequential snow and bark beetle disturbances. Forest Ecology and Management, 619, 124074.

Spatial distribution of forest damage records and sequential disturbance pathways in Finland. Panel (a) shows the distribution of individual wind, snow, and bark beetle damage polygons. Panel (b) shows locations where wind damage was followed by bark beetle damage, and panel (c) shows locations where wind damage was followed by snow damage in a subsequent year. Panel (a) is displayed as polygons, whereas in panels (b) and (c), points represent the locations of sequential disturbance records included in the analysis.

These findings are part of a broader pan-European research effort aimed at developing more resilient forests and supporting climate-change adaptation across Europe, funded by the European Union's Horizon Europe Programme through the Eco2Adapt project and by the Strategic Research Council of Finland through the ForTran project.