Resumen de Biomarker proxies as quantitative indicators of climate variability in the montcortès lake
The main objective of this dissertation was to validate and undertake the application of biomarker climatic proxies in the context of a Mediterranean lake basin. To this end, the work consists of the detailed observation of modern biomarker production, transport, deposition processes and comparing them with instrumental records on seasonal and interannual scales, exploring how the spatial and temporal patterns of variability of the biomarker signals integrated paleoenvironmental signals in the sedimentary record. The proxies studied are derived from n-alkyl and GDGT lipids. The site selected, Lake Montcortés in the Catalan Pre-Pyrenees, has been proven to be in previous studies a site sensitive to climate changes driven by Atlantic and Mediterranean influences and its sediments contain unique high resolution sequences deposited in the 20th century. In order to identify the likely sources of terrigenous biomarkers in Lake Montcortés sediments, and to study their spatial and temporal (monthly/seasonal) variability, we collected a number of plant specimens and soil samples across the catchment over a year. Our results suggest that inputs of alkyl lipids differ among particular plant groups (woody plants, grasses, emergent and submerged macrophytes), and can be distinguished from their distributions and various indices. Our study does not provide evidence for seasonal effects on the abundance and proxies of n-alkanes and GDGTs in soils. While in the suspended and settling particulate matter in the water column, significant seasonal signals were observed from the concentration and flux of lipid biomarkers due to its production, transport as well as lake circulation in different seasons, e.g. C29 n-alkane abundance in the sediment trap was correlated to local precipitation/soil runoff (especially the storm) or wind. The compound specific δ13C values of long-chain n-alkanes remained constant throughout the year in soils and sediments to faithfully preserve the initial δ13C signals of dominant C3 higher plants. The variability of GDGT-derived TEX86, temperature estimates were generally small on a short-time scale (monthly or seasonally) within the same soil type, but there was significant heterogeneity among catchment soils in the abundance and distribution of GDGTs, which might be attributed to differences in pH and soil humidity. TEX86- temperature estimates from the water column and sediment trap showed significant seasonal changes which were similar to the CTD-measured temperatures in the upper water column of Lake Montcortés.
However, in the sediment core, TEX86-estimated temperatures were potentially biased towards cooler temperatures due to contributions from methanogenic Archaea, or towards warmer signals due to allochthonous inputs from catchment soils. GDGT-MBT/CBT temperature estimates showed a large heterogeneity in the different catchment soils, and together with other evidence leads us to conclude that MBT/CBT as a temperature proxy is not applicable in the Lake Montcortés on decadal time scales.
Based on instrumental records and fractional abundance of GDGTs, a new index was used to estimate the mean annual air temperatures, which significantly recorded the onset of the Little Ice Age since the end of 15th century in the study area. The results of this thesis in Lake Montcortés can probably be extrapolated to similar lacustrine settings with caution. Thus, this thesis is also an example of the need to appraise at a catchment level the fidelity of biomarker proxies to reconstruct climate quantitatively in lacustrine sediments.
