ABSTRACT
A sound knowledge of paleovegetation is fundamental to better understanding of past climate variability, and provides necessary background for improving predictions of future climate changes. In the present study, stable isotopes (δ13C and δ15N) and lignin biomarker of sedimentary organic matter from different lacustrine deposits (e.g., tropical, subtropical and temperate lakes) were determined to reconstruct paleovegetation and to find vegetation-climate relationships.
The stable isotope signatures in lacustrine deposits of temperate (Lake DaBuSu, 44°49´ N; 123°40´ E, northeast China) and sub-tropical (Lake Erhai, 25°50′ N; 100°10′ E, southwest China) areas suggested that the organic matter in both lakes mostly originated from C3 plants throughout the depositional periods. Compositions of the alkaline CuO-oxidation products of lignin showed a significant change in terrestrial vegetation from gymnosperm-rich to angiosperm-rich plants in both lake catchments in the early-Holocene.
In the application of CuO-oxidation products of lignin, an index called lignin phenol vegetation index (LPVI) was newly developed in the present study. The LPVI of Lake Erhai and Lake DaBuSu sediment cores provided better accuracy of the vegetation changes than the previously used lignin phenol parameters, and detected subtle changes in terrestrial vegetation during the Holocene. The LPVI of both sediment cores suggested a direct link between vegetation changes and Asian monsoon variability, while the vegetation changes in Lake DaBuSu catchments did not necessarily coincide with the record from the Lake Erahi. This regional difference might result from the asynchronous nature of Asian monsoon in temperate and sub-tropical areas.
Signatures of stable carbon isotope, elemental carbon (soot) and lignin biomarker in the sedimentary core from tropical wetland (Rawa Danau, 06° 11′ S, 105° 59′ E, west Java, Indonesia) consistently indicated that there was periodic drought and precipitation variability during mid-late Holocene. Forest fire was an important factor that influenced the tropical vegetation changes during prolonged drought. The periodic drought and precipitation variability and subsequent vegetation changes in tropical wetland were found to be related to long-term environmental changes induced by El Niņo/Southern Oscillation frequency during mid-late Holocene.
The sediment core from Rawa Danau wetland was mainly composed of woody peat, whereas monosaccharides from aquatic organisms were significantly found in some sections, indicating that the carbohydrate composition in sedimentary organic matter related to changes in sources and depositional environments. Although sources and dynamics of carbohydrates are complex, principal component analysis suggested a way to distinguish terrestrial plant sources from non-terrestrial plant sources, e.g., aquatic organisms including phytoplankton, zooplankton and bacteria.
Isotopic and lignin phenol compositions of modern pollen were determined to measure a possible distortion of biomarker inferences due to the presence of fossil pollen grains in lacustrine deposits and to seek a possible use of pollens as a natural archive for isotope and lignin biomarker signatures. Exceptionally high concentrations of cinnamyl phenols were found in both raw pollen and chemically isolated sporopollenin from angiosperm and gymnosperm plants. These data implied that if pollen flux was sufficiently high, the inferences based on sedimentary lignin biomarker might be distorted. In the present study, however, reconstructions of paleovegetation using lignin biomarker were not distorted due to insignificant contribution of lignin derived from fossil pollen.
The overall results of the present study provided an integrated history of paleovegetation changes in tropical, sub-tropical and temperate areas and its linkage to regional and global climate changes. This study also demonstrated that lignin molecular stratigraphic record, in association with carbon isotope, is a powerful tool for reconstructing paleovegetations, and newly developed LPVI can semi-quantitatively interpret the biomarkers-vegetation-climate relationships.