A Critical Evaluation of Modern and Paleogene GDGT Distributions : Implications for the Application of the TEX86 Paleothermometer
Abstract
The application of glycerol dialkyl glycerol tetraether (GDGT) based sea surface temperature (SST) proxies (e.g. TEX86) has brought about major advances in our understanding of ancient climates. However, the most recent core-top calibration exercise revealed an apparent division between sedimentary GDGT distributions from cooler, (sub)polar oceans and warmer tropical oceans. As such, separate calibrations have been suggested for the reconstruction of cold (TEX86L) and warm (TEX86H) SSTs. In fact, while TEX86H retains the original TEX86 ratio of GDGTs, TEX86L utilises a different ratio. This suggests fundamentally different GDGT distributions in those environments where one calibration is favoured over the other. Ultimately, the application of either calibration to GDGT distributions in modern tropical sediments (SST >15°C) usually results in similar SST estimates, as opposed to the (sub)polar (SST <15°C) subset in which TEX86L generally produces cooler SST estimates than TEX86H. Here we investigate the relationship between the two calibrations by interrogating a compilation of Paleogene GDGT distributions, and the modern core-top calibration data-set. We find that large offsets between TEX86H and TEX86L are reconstructed across the entire SST range when considering both modern and Paleogene data sets; i.e. large offsets between TEX86H and TEX86L even where both calibrations estimate SSTs above 15°C. Similarly we also note that some (sub)polar GDGT distributions do not exhibit large offsets between TEX86H and TEX86L reconstructions, despite the cooler SSTs. In post-K/Pg sediments at mid-Waipara River, NZ, we observe a dramatic reversal in the offset between TEX86H and TEX86L (i.e. TEX86L reconstructs warmer SSTs than TEX86H). This is particularly unusual as other evidence suggests cooling through the interval; the expectation, based on the trends in the modern data set, would be a divergence in the offset between TEX86H and TEX86L as SST cools. We interrogate this offset between TEX86H and TEX86L in other modern and ancient settings in terms of GDGT distributions. Specifically, the offset is associated with low GDGT-2 to GDGT-3 ratios, and is not limited to cold regions. Moreover, GDGT distributions which exhibit low GDGT-2 to GDGT-3 ratios are associated with relatively shallow (epipelagic to upper bathyal) depositional environments. Re-evaluation of GDGT distributions from suspended particulate matter indicate that ratios of GDGT-2 to GDGT-3 increase with depth. As such, low ratios are associated with shallow water settings and upwelling zones. Together these investigations reflect depth-dependent differences in GDGT distributions, with some reflecting primarily shallow water signatures and others reflecting integration over a greater depth range. We conclude that the offset between TEX86H and TEX86L is not driven exclusively by temperature, but reflects a separate depth-dependent control on the GDGT assemblages exported to sediments; i.e. GDGTs are exported from throughout the water column and record a depth-integrated signal that can vary depending on biological and physical oceanographic features.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2012
- Bibcode:
- 2012AGUFMPP31F..04T
- Keywords:
-
- 0424 BIOGEOSCIENCES / Biosignatures and proxies;
- 0473 BIOGEOSCIENCES / Paleoclimatology and paleoceanography;
- 1055 GEOCHEMISTRY / Organic and biogenic geochemistry;
- 4954 PALEOCEANOGRAPHY / Sea surface temperature