Differences in plant-dispersal mechanisms between contrasting Brazilian savanna habitats

ABSTRACT

Background

By favouring long-distance dispersal, anemochory is often associated with open and patchy habitats, whereas zoochory enables short-distance dispersal and prevails in closed and extensive habitats. Brazilian savanna (Cerrado) on shallow rocky soils (RS) have open vegetation and are patchily distributed, whereas savanna on deep soils (DS) have dense vegetation and occur in large stands. Thus, we predicted that zoochory would be favoured in DS and anemochory would characterise RS.

Aims

We tested the hypothesis that vegetation structure and distribution patchiness are related to the characteristic dispersal mechanisms associated with closed and open habitats.

Methods

We compared the abundance of dispersal mechanisms between DS and RS with generalised linear mixed models and used an indicator species analysis to identify species associated with each vegetation type.

Results

There was no difference between DS and RS in the number of species associated with one or the other dispersal mechanism. We found fewer-than-expected zoochorous individuals and more-than-expected anemochorous individuals in RS, and more zoochorous indicator species in DS.

Conclusion

Habitat patchiness and openness are related to the composition of dispersal mechanisms. Open vegetation on shallow rocky soils favours long-distance dispersal relative to zoochory. This implies reduced plant establishment and increased leaflessness, being harsher on animal dispersers. Protected areas are often concentrated in RS, whereas DS are rapidly converted into croplands due to their excellent aptitude for mechanised agriculture. Conserving DS and RS requires a balanced complementary approach, that ensures that all landscape elements are adequately represented in protected areas.

KEYWORDS:

• Biodiversity conservation
• Cerrado; dispersal syndromes
• habitat patchiness
• plant-dispersal relationship

Acknowledgements

We thank two anonymous reviewers and the handling editor for their insightful suggestions and guidance, which greatly improved the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplementary material

Supplemental data for this article can be accessed online at https://doi.org/10.1080/17550874.2022.2132543.

Additional information

Funding

This study was partially funded by the CAPES (Coordenação de Aperfeiçoamento de Pessoal de Nível Superior, Brazilian Ministry of Education) - Finance Code 001 and by the CNPq (Conselho Nacional de Desenvolvimento Científico e Tecnológico), grant no. 430149/2018-2. The authors are grateful to CNPq for the support of doctoral fellowships to Nadjarriny and Ana Clara Abadia, and productivity scholarships to Eddie Lenza (grant no. 308236/2017-3) and José Roberto Rodrigues Pinto (grant no. 307340/2017-1). GRC thanks Coordenação de Aperfeiçoamento de Pessoal de Nível Superior (CAPES), Conselho Nacional de Desenvolvimento Científico e Tecnológico (CNPq), Fundação de Apoio à Pesquisa do Distrito Federal (FAPDF), and the USAID’s PEER program under cooperative agreement AID-OAA-A-11-00012 for financial support.