Regional vegetation changes in the tropical Andes expressed as arboreal pollen (AP%), standardised ordination scores (DCA z-scores), and aquatic vegetation ratios (D/SS); files used for plots

We compare eight pollen records reflecting climatic and environmental change from the tropical Andes. Our analysis focuses on the last 50 ka, with particular emphasis on the Pleistocene to Holocene transition. We explore ecological grouping and downcore ordination results as two approaches for extracting environmental variability from pollen records. We also use the records of aquatic and shoreline vegetation as markers for lake level fluctuations, and precipitation change. Our analysis focuses on the signature of millennial-scale variability in the tropical Andes, in particular, Heinrich stadials and Greenland interstadials. We identify rapid responses of the tropical vegetation to this climate variability, and relate differences between sites to moisture sources and site sensitivity.

Pollen abundances, and temperature and precipitation reconstruction for the Eocene to Oligocene transition in northern high latitudes

A profound global climate shift took place at the Eocene-Oligocene transition (~33.5 million years ago) when Cretaceous/early Palaeogene greenhouse conditions gave way to icehouse conditions (Zachos et al., 2001, doi:10.1126/science.1059412; Coxall et al., 2005, doi:10.1038/nature03135; Lear et al., 2008, doi:10.1130/G24584A.1). During this interval, changes in the Earth's orbit and a long-term drop in atmospheric carbon dioxide concentrations (Pagani et al., 2005, doi:10.1126/science.1110063; Pearson and Palmer, 2000, doi:10.1038/35021000; DeConto and Pollard, 2003, doi:10.1038/nature01290) resulted in both the growth of Antarctic ice sheets to approximately their modern size (Coxall et al., 2005, doi:10.1038/nature03135; Lear et al., 2008, doi:10.1130/G24584A.1) and the appearance of Northern Hemisphere glacial ice (Eldrett et al., 2007, doi:10.1038/nature05591; Moran et al., 2006, doi:10.1038/nature04800). However, palaeoclimatic studies of this interval are contradictory: although some analyses indicate no major climatic changes (Kohn et al., 2004, doi:10.1130/G20442.1; Grimes et al., 2005, doi:10.1130/G21019.1), others imply cooler temperatures (Zanazzi et al., 2007, doi:10.1038/nature05551), increased seasonality (Ivany et al., 2000, doi:10.1038/35038044; Terry, 2001, doi:10.1016/S0031-0182(00)00248-0) and/or aridity (Ivany et al., 2000, doi:10.1038/35038044; Terry, 2001, doi:10.1016/S0031-0182(00)00248-0; Sheldon et al., 2002, doi:10.1086/342865; Dupont-Nivet et al., 2007, doi:10.1038/nature05516). Climatic conditions in high northern latitudes over this interval are particularly poorly known. Here we present northern high-latitude terrestrial climate estimates for the Eocene to Oligocene interval, based on bioclimatic analysis of terrestrially derived spore and pollen assemblages preserved in marine sediments from the Norwegian-Greenland Sea. Our data indicate a cooling of ~5 °C in cold-month (winter) mean temperatures to 0-2 °C, and a concomitant increased seasonality before the Oi-1 glaciation event. These data indicate that a cooling component is indeed incorporated in the d18O isotope shift across the Eocene-Oligocene transition. However, the relatively warm summer temperatures at that time mean that continental ice on East Greenland was probably restricted to alpine outlet glaciers.

(Table 1) Pollen assemblage at DSDP Holes 96-603B and 93-605

The analysis of 104 core-catcher samples from Site 603 resulted in a continuous palynological record from the middle Miocene to early Pleistocene. Two palynological zones could be established (Zone 2, below 380 m: high Pinus, Quercus, and Carya values; Zone 1, above 380 m: high Pinus, and Sphagnum values). The marine pollen record was found to reflect the history of the vegetation of the eastern United States, the flora from Zone 2 indicating a warmer climate than that from Zone 1, which depicts a climatic cooling trend. The onset of Western Boundary Undercurrent deposition had no noticeable effect on the pollen distribution in the marine sediments.

Pollen from multicores surface samples from Namaquan mudbelt

The distribution of pollen in marine surface sediments offshore of the west coast of South Africa has been investigated to aid in the interpretation of marine pollen records of onshore vegetation changes. A transect of sediment surface pollen samples retrieved from the Namaqualand mudbelt from just south of the Orange River mouth (29°S) to St Helena Bay (33°S) indicates distinctive pollen spectra reflecting vegetation communities on the adjacent continent. Pollen concentration increases southwards, partly in relation to greater pollen productivity due to higher biomass and density of fynbos vegetation and of sedimentary processes and low pollen concentrations consequent to dilution with silt and clay from the Orange River. The distribution of specific pollen taxa suggests that the Orange River is a major contributor of pollen to the northern mudbelt declining southwards, while the pollen distribution in the central mudbelt is largely attributable to seasonal inputs of pollen from offshore berg winds and local ephemeral Namaqualand rivers. The typical fynbos elements dominate in the southern mudbelt indicating a pollen source mainly in the fynbos vegetation types. These conclusions support a companion analysis of fossil pollen records of two marine sediment cores from the northern and southern mudbelt respectively. This study demonstrates that pollen records from marine sediment cores in the Namaqualand mudbelt have the potential to be a tool to reconstruct palaeovegetation on the adjacent continent. However, to better reconstruct the palaeoclimate of South Africa and fully understand the relations between terrestrial and marine deposits, more marine surface sediments along the western coast of South Africa as well as more terrestrial surface sediments need to be studied.