Growing concerns about climate change and the increasingly tight time window for action require significant reductions in carbon emissions and large-scale removal of carbon dioxide (CO2) from the atmosphere. Natural climate solutions, such as forest restoration covers, are considered cost-effective and scalable methods of carbon removal. However, carbon removal rates can vary significantly depending on location and forest age, meaning that newly emerging forests may not provide significant carbon removal for years. New study published in Nature Climate Change sheds light on this variability and highlights the unexpected importance of protecting existing young secondary forests.

A more detailed understanding of forest growth

Traditional approaches often focus on planting new trees, but the resources for such large-scale planting are insufficient. Instead, greater emphasis is needed on natural forest regeneration on cleared areas, which can be very effective in sequestering carbon and restoring biodiversity. Existing estimates of potential carbon removal by natural forest regrowth have failed to adequately capture spatial and age-specific variations in stand age. For example, the IPCC default rates distinguish only two age classes of secondary forests: young (≤20 years) and old (21–100 years), at the continental and ecological zone levels.

To overcome these limitations, a new study mapped the density of aboveground living carbon (AGC) over time in stands aged 1–100 years, using eight times more field data than previous efforts (109,708 plots instead of 13,112). The data were grouped into 5-year age classes and combined with 66 global environmental variables that included climate, soil properties, radiation, topography, and biomes. Random forest models were trained on these data, and then a Chapman–Richards (CR) function was calculated for each grid cell, which helped smooth the estimates and more accurately reflect the natural course of forest maturation. This methodology allowed for the creation of detailed maps of the CR curve parameters, which show the maximum potential carbon density and growth rate.

Key findings on optimal age and performance

The study yielded crucial insights into the dynamics of carbon removal:

• Maximum carbon removal rates varied worldwide by up to 200-fold during the first 100 years of growth.
• Highest removal rates were estimated for forests aged approximately 20 to 40 years. Specifically, most forested ecoregions (84 %) reach their maximum carbon removal rates precisely in this age range. Annual carbon removal rates typically start low, increase, and then decline in older forests.
• Geographic variability: The highest average maximum rates are achieved in tropical and subtropical moist deciduous forests (1.57 MgC ha−1 yr−1 at age 23 ± 7 years), while Mediterranean forests show the lowest rates.

Why are existing young secondary forests so important?

Despite the usual emphasis on establishing new forests, the study revealed that protecting existing young secondary forests can provide up to 8 times more carbon removal per hectare compared to new regeneration. An optimized 25-year period for each grid cell (average 19–43 years) could remove an average of 20.7 ± 13.1 MgC ha−1, compared to 18.8 ± 12.4 MgC ha−1 during the first 25 years of new regeneration. Some sites showed increases of up to 820 %.

This potential is critical because young forests are often at risk. For example, in the Brazilian Amazon, half of secondary forests are cleared within 8 years of their establishment, while in the humid forests of Costa Rica, the average age of logging is 20 years. The study showed that an 8-year-old forest in the Brazilian Amazon would remove 36 % more carbon by 2030 than newly established stands.

The study also highlights the urgency of action: if regeneration were to start in 2025 on 800 Mha of identified forest area, up to 20.3 billion MgC could be removed by 2050. A delay of 5 or 10 years reduces this potential by about a quarter to a half. This highlights the critical importance of rapid action.

Policy challenges and recommendations

Protecting young secondary forests is vital, but there are limited mechanisms to support them. For example, current carbon market methodologies do not recognize the protection or improved management of very young secondary forests. Regulations require a minimum of 10 years of cleared land before afforestation projects, and forest protection projects require forests to be at least 10 years old. These barriers prevent the effective integration of young secondary forests into climate strategies. The study also found that, compared to the IPCC default rates, their modeled rates are 26 % lower in forests younger than 20 years and 18 % higher in forests aged 21–100 years, with IPCC estimates failing to capture 1.2- to 15.8-fold variation across ecological zones.

Therefore, it is key:

• Prioritize the protection of endangered young secondary forests alongside older stands with large carbon stocks.
• Rethinking current carbon marketsto include mechanisms for the protection and management of young secondary forests.
• Take into account the socio-economic context and the role of local communities when implementing nature-based solutions. Climate finance must take into account the potential adverse impacts on people's livelihoods.

Significance and future directions

This study provides more detailed and accurate estimates of carbon removal rates, allowing for more targeted optimizations. It shows that protecting and restoring forest cover is essential to achieving the goals of the Paris Agreement. Although the model has some limitations, such as a bias in the data towards northern temperate forests and a failure to account for future climate change, it provides invaluable information for policymakers and project developers. By strengthening global forest carbon sinks – protecting intact forests, maintaining secondary forests for immediate removal, and supporting new forests for future gains – we can more effectively mitigate climate change. Spring

Glossary of key terms

• Secondary Forest: A forest that grows naturally on land that was previously deforested or disturbed by human activity (e.g. logging, agriculture) or natural events (e.g. fire).
• Carbon Removal: The process of removing atmospheric carbon dioxide (CO2) and storing it in terrestrial, oceanic, or geological reservoirs. In the context of forests, it refers to the sequestration of carbon by trees.
• Aboveground Live Carbon (AGC): The total amount of carbon contained in the living biomass of trees above ground (trunks, branches, leaves). It is measured in MgC ha−1 (megagrams of carbon per hectare).
• Chapman-Richards (CR) function: A nonlinear growth function commonly used in forestry to model typical nonlinear forest growth, which includes phases of initial slow growth, accelerated growth, and subsequent deceleration/saturation.
• Natural Climate Solutions (NCS): Measures to protect, restore and improve land management that increase carbon sequestration and/or reduce greenhouse gas emissions.
• IPCC Tier 1 default rates: Standardized, rough estimates of rates of biomass and carbon stock change provided by the Intergovernmental Panel on Climate Change (IPCC) for different forest types and ages at the continental and ecological zone levels are less spatially and temporally specific.
• Megagram of carbon per hectare per year (MgC ha−1 yr−1): A unit for measuring the annual rate of carbon removal, i.e. the amount of carbon (in megagrams) sequestered per hectare per year.
• Additionality: A principle in carbon offsets that requires that the emission reduction or carbon removal be additional to what would have occurred without the project or intervention in question.
• Durability: A principle in carbon offsets that concerns long-term carbon storage, i.e. ensuring that the carbon sequestered by a project remains stored for decades to centuries and is not released back into the atmosphere.
• Ecoregion: A large area of land or water containing a geographically distinct set of natural communities and environmental conditions.
• Root Mean Square Error (RMSE): A statistical method for measuring the average size of a model's prediction errors, which shows how close the predictions are to the observed values.
• Coefficient of Determination (R2): A statistical measure that represents the proportion of the variance in a dependent variable that can be predicted from the independent variable (the variable factors in the model). A value closer to 1 indicates a better fit of the model.
• Net Zero Targets: Targets that countries, companies, or organizations set to achieve a balance between the amount of greenhouse gases released into the atmosphere and the amount removed.
• Random Forest models (Random Forest Models): A machine learning algorithm that works by creating a large number of decision trees and the output is a class, which is the mode of the classes (for classification) or the mean prediction of the individual trees (for regression).
• Environmental Covariates: Environmental variables (e.g., climate, soil properties, topography) that are used in statistical models to explain or predict changes in another variable.