Format: ORAL

Carlos A. Ordez-Parra1,2, Lara Amaral-Garcia2, Rafael Figueiredo2, Fernando A. O. Silveira2

1. Programa de Pós-Graduação em Biologia Vegetal, Departamento de Biologia Vegetal, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil. 2. Centro de Síntese Ecológica e Conservação - CSEC, Departamento de Genética, Ecologia e Evolução, Instituto de Ciências Biológicas, Universidade Federal de Minas Gerais, Belo Horizonte, Brazil.

Seed-based conservation and restoration are increasingly recognised as cost-effective strategies to protect biodiversity and restore degraded ecosystems. Both strategies depend on seeds ability to remain viable in medium- to long-term storage, which, in turn, is directly related to seeds desiccation tolerance a trait whose assessment is highly time-consuming and challenging to scale to various species. To overcome this shortfall, trait-based models to predict desiccation tolerance using easy-to-measure traits, such as seed mass and moisture content, have been proposed. Here, we used these models to explore the potential of seed conservation in two hyperdiverse yet threatened open ecosystems in Brazil: rock outcrop vegetation and the Cerrado, with the ultimate goal of delivering knowledge about native seed conservation and building capacity for their use in conservation and restoration projects. For rock outcrop vegetation, we built a database of 16 seed traits for 383 species. Unfortunately, data on desiccation tolerance was available for only 34 species, all of them producing orthodox seeds. Still, the available information for seed mass and moisture content for 87 species indicates a high potential for seed conservation, as suggested by the few empirical studies available for these and similar ecosystems worldwide. An ongoing project in the Cerrado is assessing seed desiccation tolerance in 84 species 56 of them with unknown storage behaviour currently commercialised by local communities in Central Brazil. This species pool includes groups whose seed desiccation tolerance has been little studied, such as perennial grasses, which are a fundamental element of Cerrado plant communities. Ongoing data collection will be compared with the predictions of global trait-based models to assess whether these models produce accurate predictions of seed desiccation tolerance and, thus, used as a tool to facilitate the seed conservation work currently carried out by local communities.