Ecosystem Restoration Should Rely on Self-Repairing Mechanisms

Abstract

One third of soils globally are considered degraded, due to the effects of climate change and inappropriate land use, which drives a large research interest in practices that aim for recovery of soil health. Soil degradation often translates into a loss of soil organic matter (SOM) and soil structure, resulting in reduced physical, chemical, and biological soil functions, such as water storage and infiltration, nutrient storage and cycling, as well as habitat provisioning and related biodiversity support. In many cases, stabilization of SOM and soil structure is closely interlinked through physical protection of SOM in soil aggregates. They form when organic gluing agents bind to soil minerals, and when finally microaggregates (diameter < 250 μm) are enmeshed by fungal hyphae and roots to form macroaggregates (diameter > 250 μm), which can also include sand particles. Macroaggregates have a turnover time of weeks to years and are sensitive to land management, whereas microaggregates have a turnover time of decades (von Lützow et al. 2008). Therefore, SOM re-stabilization in soil aggregates does not only contribute to the recovery of the physical structure of the soil, but also to overall soil health and storage of carbon on the scale of years to decades.

Many studies have demonstrated a possible recovery of SOC, which is the main component of SOM, and soil structure in croplands after topsoil loss, conservation agriculture practices, or conversion to permanent vegetation like grasslands (Burger et al. 2023; Kösters et al. 2013). Yet, soils with permanent vegetation like grasslands or even forests can also be degraded due to non-adapted land use, climate, and ecological change, which then also results in a loss of SOC and soil structure (Tian et al. 2023). Unfortunately, available studies that focused on SOC recovery of degraded grassland and forest soils showed varying results (Tian et al. 2023). Global Change Biology recently published a meta-analysis that examined the long-term effects of active restoration and natural regeneration practices in grasslands, shrublands, and forest soils. The studies summarized for the first time how these practices affect SOC stocks and the SOC stored within different aggregate fractions (Ye et al. 2025). The authors compiled 94 observational studies from around the globe that measured long-term effects of either active restoration or natural regeneration, as well as those that compared both. Long-term here means a restoration time of up to 40 years for grasslands and up to 60 years for forests. Natural regeneration involved minimal human intervention like fencing and conservation or grazing exclusion. Active restoration involved more human intervention like afforestation, planting, seeding, and reforestation. In addition, the authors also draw specific attention to soil aggregate size fractions, as well as to soils with initial low or high SOC content (< 1% or > 1% SOC, respectively) as well as subsoil carbon.

This meta-analysis is the first study to demonstrate that natural regeneration outcompeted active restoration in terms of SOC accumulation and accumulation of SOC in aggregates in ecosystems with low management intensity (forests, grasslands, shrublands) in the long term (> 40 years). In forests with natural regeneration, there was initially a decrease in microaggregates, but this effect disappeared 20 years after the start of regeneration. Additionally, the bulk SOC recovered after 40 years, as did the SOC in both the macro- and microaggregates, as well as in the silt and clay fractions. This indicates a recovery of soil structure (Ye et al. 2025). The ecological benefits of natural regeneration, such as on the diversity of plant functional types, likely enabled positive feedback loops that increased soil fertility, supported C cycling, and long-term SOC accumulation and aggregate formation (Weiskopf et al. 2024). The main environmental factors considered in this study that contributed to the recovery of bulk and macroaggregate SOC were the soil depth (topsoil or subsoil) and initial SOC content of the soil. Areas with low initial SOC content showed the largest increases in SOC content under natural regeneration; these soils were seen as severely degraded and thus had a great potential for recovery (Ye et al. 2025). These findings are remarkable since the restoration of such sites is slow and challenging and requires a deep understanding of site characteristics; also, secondary successions and the effects of pioneer species are crucial to the success of restoration practices (Chazdon et al. 2024).

In the subsoil, the positive effects of natural regeneration on bulk SOC stocks and the SOC stocks in aggregates were also visible, whereas those of active restoration were not as clear. This was especially visible in the bulk soil and in the silt and clay fraction, where subsoil accumulated more SOC under natural regeneration than under active restoration (Ye et al. 2025). Since subsoil is not always a (feasible) target of active restoration practices in grass-, shrubland, and forests, natural regeneration practices can enhance the recovery of subsoil SOC accumulation and soil function due to greater root development and plant diversity (Quartucci et al. 2023).

Although the study by Ye et al. (2025) shows that natural regeneration is outperforming active restoration practices, this does not always mean that degraded areas should be fenced and that SOC will then recover within 60 years without intervention or assistance. To utilize the self-repair mechanisms of ecosystems, it is important that former degradation is stopped before it changes by regional climate change and species composition in an irreversible manner. Therefore, it is advised to determine in what way the soil functions are lost by degradation, and how they can be supported to make use of the self-repairing mechanism of the soil (Chazdon et al. 2024). Another type of soil that could be focused on is agricultural soils, which are often seen as having high potential for SOC recovery, due to their lower SOC contents from intensive use (Tian et al. 2023). However, agricultural soils also need to provide food security, which might make natural regeneration not always suitable or possible in every region, and active restoration strategies have to navigate possible trade-offs (Moinet et al. 2023). Future research could focus on degraded arable land and whether natural regeneration into grassland, shrubland, or forest would be a viable option. If natural regeneration would not be an option, research could focus on which active restoration practices could increase SOC in soil but also enhance the synergies to sustain this SOC level and not compromise yield.

In general, both natural regeneration and active restoration practices showed long-term increases in bulk soil SOC and macroaggregate SOC (Ye et al. 2025), that is, both practices are a valuable tool to sequester C in soils, and, thus, to contribute to climate change mitigation. However, the study by Ye et al. (2025) also shows that these practices are particularly successful if they enhance the ecosystem's functioning and self-repair mechanisms. If this is granted, recovery of biodiversity, SOC, and soil structure go hand in hand (Weiskopf et al. 2024), and shows that respective restoration strategies should be site-specific (Tian et al. 2023). The recently published recommendations by the G20 conference in Brazil call for restoration of degraded ecosystems, to support climate change adaptation, improve livelihoods, and meet the sustainable development goals (G20 Global Land Initiative 2024). Therefore, we need to focus on the ecological benefits of regeneration, particularly those that promote synergies in the soil, to ensure that restoration is economically viable and contributes to long-term improvement in soil and ecosystem functioning. The study of Ye et al. (2025) shows that this is feasible when considering the interplay of SOC accrual and soil structure restoration, whereas at the same time calling for holistic approaches to active restoration practices to better facilitate the ecosystem's self-repairing mechanism.

The author declares no conflicts of interest.

This article is a Invited Commentary on Yuqian Ye et al., https://doi.org/10.1111/gcb.70255.