ALBERT

Description: Radiocarbon (14C) ages obtained from foraminifera calcite are used for paleoseawater reconstruction and carbon cycle dynamics for the past 30,000 years. Up to now, the effects of diagenesis on this proxy were poorly understood. The aim of this master thesis is to investigate if systematic differences in radiocarbon ages in pristine and diagenetic overprinted benthic and planktic arctic foraminifera exist. The foraminifera shells were obtained from the sediment core PS92/93-2 close to the Yermak Plateau in the Eurasian Basin. Optical microscopy, scanning electron microscopy and energy dispersive X-ray analysis were used to check for diagenetic alteration. Authigenic calcite precipitation on planktic arctic foraminifera (N. pachyderma sin.) were observed in four intervals (between 15–230 cm bsf) and three types of overgrowth are defined for classification. Radiocarbon measurements are done on untreated pristine shells and shells with overgrowth. The results of this study demonstrate that secondary calcite overgrowth, precipitated on foraminifera shells, causes systematic younger 14C-ages. Leaching experiments are done on planktic foraminifera shells which show tall overgrowth crystals. With leaching of the outer shell and the measurement of the radiocarbon content of both the leachate and the leached sample it is possible to identify the effect of the overgrowth. The main outcome of the leaching experiment is that the leachates are composed of the 14C signal of the shell, 14C-enriched overgrowth, and adsorbed atmospheric CO2. The source of the overgrowth needs to be determined by further research. The stable isotopic composition of pristine foraminifera and foraminifera with overgrowth reveal no significant offset. It is likely that the calcite precipitated from 14C-enriched pore water.

Description: The TEX86 paleothermometer has been extensively used to reconstruct past sea water temperatures, but it remains unclear which export depths the proxy represents. Here we used a novel approach to better constrain the proxy recording depths by investigating paleotemperature proxies (TEX86, UK'37, RI-OH and RI-OH') from two pairs of proximal (〈 12 km apart) cores from Chilean and Angola margin, respectively. These cores are from steep continental slopes and lower shelves, which leads to a substantial difference in water depth between them despite being closely located. Surprisingly, the deep and the shallow UK'37 records at the Chilean margin shows dissimilarities, in contrast to the similar records from the Angola margin, which may be due to post-depositional alteration at the former sites. In contrast, the TEX86 records were statistically indistinguishable between the sites at both the locations, even though the GDGT [2]/[3] ratio suggests GDGTs derived from potentially different archaeal communities residing at different depths. A short-lived difference between the TEX86 records is observed during the last glacial period at the Angola margin, possibly due to a contribution of Antarctic Intermediate Waters to the deep site. Modelling suggests that the TEX86 source signal at our core sites reaches its peak abundance at water depths shallower than 350 m. The RI-OH and RI-OH' records show similar variability as the TEX86 records, although regional differences in their absolute temperature estimates exist. Our approach using proximal sediment cores at steep slopes appears useful to constrain the export depth of organic proxy signals for paleoreconstructions.

Description: To compare the neighboring cores ODP Leg 202 Site 1235 and Site 1234, ca. 12 km apart from each other, we constructed an age-depth model with the statistical package Bacon (Blaauw & Christeny, 2011). It is based on the published benthic foraminifera radiocarbon measurements from Muratli et al. 2010 and supplemented here with three new benthic foraminifera 14C data for Site 1235 and the Laschamp event at 41.000 years BP based on correlating magnetic susceptibility (Mix et al., 2003). Radiocarbon ages were converted to calendar years using Marine09 data set (Reimer et al., 2009) with the reservoir age correction given by Muratli et al. 2010.