Current knowledge and research gaps in Cerrado seed ecology: from seed viability to ecological restoration

Carlos A. Ordóñez-Parra , Ailton G. Rodrigues-Junior, Davi N. Oliveira, Diego F. E. Escobar, Heloiza L. Zirondi, Lara Amaral-Garcia, Mariana Dairel, Mateus R. B. Bispo, Mateus C. Silva, Pedro F. Cruz-Júnior, Talita Zupo, Elson J. S. Silva, Gabriel S. T. Motta, Hudson G. V. Fontenele, Jamily S. Pereira, Joana P. B. Nascimento, Rafael Figueiredo, Raquel G. Martins, Rosani K. Reinke, Tainá C. Arreguy, Ana C. C. Oliveira, Liz D. Marinho, Luis F. Daibes

Although seed ecology underpins plant recruitment, conservation, and restoration, knowledge in this field in the Cerrado, the largest Neotropical savanna, remains fragmented across taxa, regions, environmental factors, and stages of the seed life cycle. Here we conducted a scoping review to map existing evidence, identify gaps, and synthesise ecological patterns in this megadiverse biome. We reviewed 357 studies published between 1973 and 2025, covering 644 species from 319 genera and 79 families, and characterised temporal, thematic, taxonomic, functional, methodological, and geographical research coverage. We also assembled a biome-wide dormancy profile for 1,473 species and descriptively reanalysed germination responses to light and temperature. Publication output increased markedly after the late 2000s, but research was concentrated on germination, controlled experiments, woody species, and central and southern Cerrado regions. More than half of the species were investigated within only one research topic, and laboratory–field integration was limited. Reported seed viability varied widely among families, while community soil seed banks were dominated by herbaceous species and appeared largely transient and seasonally replenished. Approximately two-thirds of species in the dormancy profile were classified as dormant, primarily with physiological or physical dormancy, although direct experimental evidence remained limited. Among the species and conditions evaluated, light generally favoured germination, and the highest germination under tested constant temperatures occurred most frequently at 25 or 30 °C. Heat shock was usually neutral or detrimental, evidence for smoke responses remained inconclusive, and responses to water availability varied with habitat, seed traits, provenance, and developmental stage. Biotic interactions were seldom studied beyond seed predation, and applied research remained poorly integrated from seed collection and storage to field establishment. Progress requires coordinated, field-validated, trait-based research linking seed traits and environmental processes across populations, vegetation types, and regeneration stages.