Portal de Periódicos da CAPES

Artigo Produção Nacional Revisado por pares

BIBTEX RIS

Changes in Holocene tree cover density in Cabo Frio (Rio de Janeiro, Brazil): Evidence from soil phytolith assemblages

2012; Elsevier BV; Volume: 287; Linguagem: Inglês

10.1016/j.quaint.2012.02.044

1873-4553

Heloísa Helena Gomes Coe, Anne Alexandre, Cacilda N. Carvalho, Guaciara M. Santos, Antonio S. da Silva, Leandro O.F. Sousa, I. F. Lepsch,

Cassava research and cyanide

The coastal region of Cabo Frio, in the Brazilian state of Rio de Janeiro, is characterized by xeric vegetation. Surrounded by humid forest, it is considered as a phytogeographical enclave. The vegetation is adapted to the dry conditions which are present locally, mainly due to the Cabo Frio coastal upwelling. Although Quaternary changes in the intensity of the upwelling were reconstructed from coastal lagoons and oceanic sediment records, the lack of continental deposits has precluded vegetation reconstructions. The objective of this study was to investigate the pedogenic features of two soil profiles developed under patches of dry forest in the Cabo Frio area and to assess their potential for inferring past changes in tree cover density, from their phytolith assemblages. For this purpose, field and petrographical observations, C content, C/N ratio, phytolith content and phytolith assemblages were investigated. Soil phytolith assemblages were compared to modern phytolith assemblages. The deepest soil organic carbon (SOC) samples were analyzed in 14C-AMS. Related 14C mean age values were interpreted as the youngest ages of the oldest SOC. These data suggest a first soil development phase, occurring after 13 ka cal BP followed by erosive and depositional episodes and by a second soil development phase. Given the complexity of the studied soil sequences, an attempt to quantify the sources of SOC and phytoliths would require investigation of many more soil samples than the ones collected. This prevents interpretation of the phytolith sequences as continuous chronological sequences. However, for both profiles phytolith indices D/P from A, Ab and bottom horizons can be compared, assuming a bicompartmental distribution of phytoliths in relation with each of the soil development phases. D/P values range from 0.8 to 4, in agreement with what would be expected for dry forest D/P values, and do not substantially change in both profiles. This comparison suggests that the tree cover density of the successive vegetation sources did not suffer considerable change over the period under analysis (last 13 ka for profile A) and never reached the tree density of the humid forest currently widespread in the Rio de Janeiro state.