Balancing natural forest regrowth and tree planting to ensure social fairness and compliance with environmental policies

摘要

1. The environmental benefits and lower implementation costs of (assisted) natural forest regrowth (NFR) compared to tree planting qualify it as a viable strategy to scale up forest restoration. However, NFR is not suitable in all places, because the potential for forest regeneration depends on the socio-environmental context and differs greatly over space and time. Therefore, it is critical to quantify the potential contribution of NFR for reaching forest restoration targets and complying with environmental policies. 2. Here, we quantify the socio-environmental consequences of NFR by considering four targets differing in restored area in the Atlantic Forest (6, 8, 15 and 22 Mha). We quantified the compliance with environmental policies, expected distribution of natural and restored vegetation within the biome and social fairness (distribution of restoration efforts and costs within small, medium and large-sized properties) of two hypothetical forest restoration scenarios. 3. We show that large-scale forest restoration prioritizing the areas with the highest potential for NFR (Scenario I) allows us to comply with one-third of the current environmental debt in the Atlantic Forest. Furthermore, this scenario disproportionately burdens specific types of land use, increases socioeconomic inequalities and concentrates restoration activities in regions in which the natural vegetation cover is already high. 4. By contrast, Scenario II-eradicating the environmental debt that results from environmental policies, then prioritizing areas with the lowest overall restoration costs until reaching the restoration targets-is socially fairer and maximizes compliance with environmental policies. Its outcomes are more homogeneously distributed among counties and small, medium and large-sized properties from the Brazilian Atlantic Forest. Despite doubling the implementation costs, the lower overall restoration costs in Scenario II result from significantly lower opportunity costs than