Deep soils found in forests may be less effective at storing carbon in the long term than previously assumed, potentially reducing the net climate benefits of tree planting, a University of Stirling professor has warned.

Professor Jens-Arne Subke of the University’s Faculty of Natural Sciences has co-authored a new commentary with Dr Thomas Parker of the James Hutton Institute that builds on recent research led by Professor Subke, which warned that the climate benefits of tree-planting could be overstated if soil carbon losses aren’t included in calculations.

Published in Global Change Biology, the commentary examines the findings of a BOKU University led study focused on European beech forests in Central Europe, which showed that ignoring deep soil carbon levels could lead to over‑optimistic estimates of how much carbon forests are able to store.

While the European study focused on beech forests, earlier research by Professor Subke’s team looked at pine plantations in Scotland – pointing to a wider pattern.

Across the world, significant tree planting campaigns are underway using photosynthesis to store carbon in wood, roots, and soil. However, Professor Subke believes the recent findings show that using forests and deposition of carbon to deep soils may not be as safe a long‑term option as previously hoped.

Professor Subke explained: “Our findings emphasised that we cannot over-rely on forests to mitigate the impacts of climate change because there is still so much that we don’t understand. Despite accumulating tree biomass, we may be losing carbon capital – the carbon stored long‑term in soils and ecosystems – to the atmosphere.”

The study, published last year, found that soils under mature pine forests had about half as much carbon as nearby soils that stayed as grassland, and that the carbon lost from the soil was equal to around a third of the carbon that the trees had absorbed from the atmosphere.

The team also found that carbon left in forest soils was less stable, meaning that it could break down and be released more easily in the future.

During the study, researchers took soil samples from 16 sites in the Scottish Lowlands where pines had been planted on former long-term grasslands - with the oldest planted 68 years ago.

These samples were analysed to assess both carbon content and stability, consistently showing that soil carbon declined as trees aged.

Dr François-Xavier Joly of the French Institute for Agriculture Research (INRAE) in Montpellier, who led the study in Professor Subke’s research team added: “There are important financial incentives for landowners to plant more trees; however, these are linked to presumed benefits brought by a change in vegetation towards forests.

“Our research has added an important aspect to these schemes by clarifying the consequences of tree planting within the soil. Administration of the Woodland Carbon Code, or equivalent schemes, must take account of potential soil losses - which we were able to demonstrate across a significant area in Scotland.”

Commentary co-author, Dr Thomas Parker, said: “Forests are an essential for human and planetary well-being for a range of reasons, but we need to acknowledge that they are not a silver bullet for all our problems. There are complexities and trade-offs that need to be understood to maximise the net benefits that we gain from forests.”

Dr Mike Perks, climate change scientist at Forest Research added: “More research is needed to better understand carbon storage in soil, including accounting for variations in soil type and texture, tree species productivity, root turnover and density. Also important is increasing our understanding of soil depth, soil processes and where soil carbon ultimately ends up.”

The commentary, Uptake and release - what is driving change in the net carbon budget in forest soils? was published in the Global Change Biology.

Temperate grassland conversion to conifer forest destabilises mineral soil carbon stocks was published in the Journal of Environmental Management.

Work took place in collaboration with Colorado State University and Forest Research and was funded by a collaborative grant from the National Environmental Research Council (NERC) and National Science Foundation (NSF).