Planting Trees

Underrepresentation of Large-seeded Tree Species Compromises the Conservation Value of Restoration Plantations

Brancalion, P. H. S., C. Bello, R. L. Chazdon, M. Galetti, P. Jordano, R. A. F. Lima, A. Medina, M. A. Pizo, and J. L. Reid. 2018. Maximizing biodiversity conservation and carbon stocking in restored tropical forests. Conservation Letters e12454.


Download PDF

Walking along restoration plantations in the Atlantic Forest of Brazil, one rapidly realize that plantings represent only a few “winner” trees, that are widely used in most projects. These species are typically produced in nurseries from seeds harvested from certain species that are preferred by seed collectors and may not represent the much larger pool of native tree species from a megadiverse tropical forest. Species selection is an important human legacy in the restoration of native ecosystems, but is still poorly understood as a driver of the conservation value of restored forests.

Ideally, the selection of native species for use in restoration should be based on the species assemblage of reference ecosystems (i.e., less disturbed ecosystems of the region representing a benchmarking for restoration targets). In the real world, anthropogenic drivers as seed cost may undermine the ecological justifications for planting a given species with unintended consequences for restoration success. Yet, the drivers and consequences of tree species selection in restoration have remained poorly explored.

To address the implications of tree species selection in the restoration of Brazil’s Atlantic Forest regarding carbon stocking and species conservation, we made use of a comprehensive dataset of seedling acquisition records from 961 restoration projects, more than 14 million seedlings, 192 forest remnants, and functional data from 1,223 tree species. Overall, small-seeded, abiotic-dispersed tree species were more common in restoration plantations than expected by their occurrence in forest remnants. This biased representation of tree species in restoration compromised the conservation value of plantations in two different ways: by underrepresentating species important for large vertebrate frugivores, and by overrepresenting species with softer wood that may have reduced potential for carbon sequestration.

The way forward will require revisiting the market structure of tree seedling production in forest nurseries,that have neglected the production of large-seeded, animal-dispersed species due to their higher seed price. Incentives for including these species could come in the form of a bonus in carbon markets, as the additional carbon credits generated by the higher carbon stocks accumulated by these species would financially compensate the higher costs of their production. By shedding light onto the drivers and consequences of tree species selection in restoration, it is possible to develop strategies for promoting the use of high-conservation value trees in restoration, and to help save species from the threat of extinction.


The PARTNERS connection
Many scientific papers (and other important human achievements) are born from bar discussions, and this was the case for this paper. After a very exciting day of discussions in a PARTNERS workshop, two PARTNERS members realized how little we know about the species assemblages that are used in restoration. Then, a task force was created joining PARTNERS members and other collaborators to develop this paper. This paper has had an impact by redefining the strategy of native seedling production in some influential NGOs working with restoration in Brazil.