Skip to main content

Sea-surface temperature evolution mapping project based on UK37 str atigraphy

Deliverables

A significant breakthrough in paleoenvironmental sciences has been the discovery in the mid-eighties that the marine fossil record of organic compounds called alkenones could provide insights into past-changes in sea surface temperature (SST). Since then, the molecular parameter UK37’ has been steadily gaining in acceptance for palaeotemperature reconstruction, and an increasing number of research groups world-wide are now measuring it. Some analytical questions, however, still need to be addressed as to date only relatively few studies have focussed on the analytical constraints to measure UK37’. From our study, it is recommended that in future, analysts: -Pay particular attention to the chromatographic analysis. -Reappraise their methods (including clean-up steps as necessary) to overcome the problem of complex matrices. -Report values or ranges of alkenone concentrations with UK37' measurements (as low concentration increases the error of analysis). -Agree on a method for absolute quantification of alkenone concentration. Adhering to these guidelines will help the organic geochemical community provide accurate data that is comparable between groups. 24 laboratories world-wide have participated in an anonymous inter-laboratory comparison study, and analysed the alkenone composition of several sediment samples and alkenone standards. Each laboratory was provided with eight test samples (five sediments and three mixtures of alkenone standards) and a code to identify itself. The alkenone standards provided were mixed in three different proportions (50/50, 25/75 and 75/25; i.e. UK37’= 0.4695, 0.726 and 0.228, respectively, calculated from the concentration of the solutions). Sediments samples were obtained from an array of marine locations, to encompass a variety of sea surface temperatures (12-27 degrees Ceslius), depositional environments (oligotrophic to upwelling), and organic composition. From this study it has been estimated that the differences between UK37’ temperature estimates, from the analysis of oceanic sediment samples, between two laboratories may differ by up to 2.1 degrees Celsius or less (i.e. 95% confidence level) owing to analytical uncertainties. In addition, the precision of the UK37’ temperature estimates from a laboratory is estimated to be up to ± 1.7 degrees Celsius or less (95% confidence level). However, this repeatability limit can be higher (more than 2 degrees Celsius) when the concentration of alkenones in the samples is at least less than 300 ng/g (dry weight). There is certainly scope, and the need, to reduce these values of reproducibility and repeatability to improve confidence in the data, not from a particular laboratory, but from the community of scientists involved in alkenone research.
Successful application of UK37 as a SST proxy requires an understanding of various factors (other than SST) which may influence the value of the parameter. These include changes in the contributing coccolithophorid species through time, water salinity, and diagenetic modification of the alkenones in the water column and sediment. A detailed comparison of the UK37’ and coccolithophorid records revealed the absence of significant changes in the UK37’ values, despite major changes in coccolithophorid assemblages. The collected evidence supports the conclusion that changes in alkenone-producing organisms in the past do not involve changes in the UK37’-temperature relationship. Evidence that diagenesis is not an important factor in altering the UK37’ signal has been obtained from the analysis of a deep-sea sediment core underlying the Benguela up welling system off SW Africa which provides a continuous time-series of sea-surface temperature (SST) for the past 4.5 m.y. The alkenone data and results from chemotaxonomic and early/pre-Quaternary studies suggest that UK37’ does not appear to be significantly affected by either Pliocene-Pleistocene species extinctions or variations in production depth, nutrients and seasonal production maxima. The suboxic sediments at Site 1084 are not typical for conditions where post-depositional alkenone diagenesis may be significant. Moreover, complete diagenetic removal of the more unsaturated alkenone has not occurred in samples from the mid-Pliocene warm period as UK37’ < 1 throughout. These findings further increase the confidence in the application of UK37' as a palaeoclimate proxy. Findings from the TEMPUS research project confirm the applicability of the UK37' parameter as a proxy for past sea-surface temperature (SST), in demonstrating the robustness of the UK37' values in response to changing coccolith (source organism) assemblages and environmental factors other than temperature. Comparisons between UK37' data and coccolithophorid records were made by counting coccoliths in order to compare their specific distribution with the alkenones from the same sediments in the tropical Indian Ocean, the tropical Atlantic Ocean, and the central North Atlantic Ocean. Comparison of the UK37’ and coccolithophorid records revealed the absence of significant changes in the UK37’ values, despite major changes in coccolithophorid assemblages. The collected evidence supports the conclusion that changes in alkenone-producing organisms in the past do not involve changes in the UK37’-temperature relationship. The relationship between SST and sedimentary UK37' values assumes no significant preferential alteration in the relative abundance of either of the unsaturated alkenones. Evidence that diagenesis is not an important factor in altering the UK37’ signal has been obtained from the analysis of a deep-sea sediment core underlying the Benguela upwelling system off SW Africa which provides a continuous time-series of sea-surface temperature (SST) for the past 4.5 m.y. The alkenone data and results from chemotaxonomic and early/pre-Quaternary studies suggest that UK37’ does not appear to be significantly affected by either Pliocene-Pleistocene species extinctions or variations in production depth, nutrients and seasonal production maxima. The suboxic sediments at Site 1084 are not typical for conditions where post-depositional alkenone diagenesis may be significant. Moreover, complete diagenetic removal of the more unsaturated alkenone has not occurred in samples from the mid-Pliocene warm period as UK37’ < 1 throughout. These findings further increase the confidence in the application of UK37' as a palaeoclimate proxy.
Estimation of past sea-surface temperature (SST) from the composition of alkenones in marine sediments (UK37) requires a calibration equation to be constructed. These exist in the literature, but part of the TEMPUS project was to produce a new calibration (based on surface sediments and modern SST), including data from areas hitherto not considered, such as the South Atlantic and some important marginal seas, to obtain a global picture of the distribution of UK37' in sediments from all oceans and marginal seas, and attest to the global application of UK37 as a proxy. Compilation of our new results with published core-top calibrations from other oceanic regions, revealed a high degree of accordance. Muller et al. (1998), therefore, established a global core-top calibration based on 370 sites between 60 degrees South and 60 degrees North in the Atlantic, Indian and Pacific Oceans: (1) UK/37 = 0.033 SST + 0.044 (n = 370, r2= 0.958). Inclusion of an additional 87 data points, produced more recently within the TEMPUS project and from the literature, yields the following, almost identical, equation: (2) UK' = 0.033 SST + 0.049 (n = 477, r2 = 0.962). Previous work from TEMPUS partners has centred on developing calibrations of UK37' and SST using sediment core-top samples, from the North-eastern Atlantic, South China Sea and the Indian Ocean. Within TEMPUS this work continued in other areas hitherto not considered, such as the South Atlantic and some important marginal seas, to obtain a global picture of the distribution of UK37' in sediments from all oceans and marginal seas, and attest to the global application of UK37' as a proxy. Sedimentary UK37' values collected within TEMPUS and data reported in the literature, were compared with modern SST values to construct an equation to calibrate the alkenone data (UK37') in terms of SST. Muller et al. (1998), therefore, established a global core-top calibration based on 370 sites between 60 degrees South and 60 degrees North in the Atlantic, Indian and Pacific Oceans: (1) UK/37 = 0.033 SST + 0.044 (n = 370, r2= 0.958). Inclusion of an additional 87 data points, produced more recently within the TEMPUS project and from the literature, yields the following, almost identical, equation: (2) UK' = 0.033 SST + 0.049 (n = 477, r2 = 0.962). The standard error of estimate for this relationship is ±0.044 UK'37 units or ±1.3 degrees Celsius. It is identical within error limits to equation 1 and also in accordance with the widely used laboratory calibrations for Emiliania huxleyi by Prahl and Wakeham (1987: UK'37 = 0.033 T + 0.043) and Prahl et al. (1988: UK'37 = 0.034 T + 0.039). Hence the calibration basis for UK'37 as paleotemperature tool becomes increasingly stable.
Climate modelling requires realistic specification of boundary conditions, including sea-surface temperature. The results of CLIMAP (1976, 1981) included a map of SST for the Earth at 18,000 years b.p. These data are still used today as boundary conditions in climate models, although there has been no concerted effort to validate this important data resource. One of the key outputs of the TEMPUS project are SST maps for much of the world's oceans for selected time periods: the last glacial maximum and the Eemian, with additional regional data for less well-defined climatic intervals. The SST data were derived from the UK37' parameter. Climate modellling requires realistic specification of boundary conditions, including sea-surface temperature. The results of CLIMAP (1976, 1981) included a map of SST for the Earth at 18,000 years b.p. These data are still used today as boundary conditions in climate models, although there has been no concerted effort to validate this important data resource. The TEMPUS project set out to construct new SST maps for key time intervals of particular climatic importance, using the UK37' proxy. One of the challenges that TEMPUS partners have had to overcome is the definition of the specific time slices or intervals. This issue is still being debated within the international community and although some consensus exists there are still some issues that need to be resolved. We have, thus, adopted two strategies to mapping, depending on the degree of consensus about the definition of a particular interval. Thus, for the intervals corresponding to the modern/Holocene, the Last Glacial Maximum and the Eemian, an average value of SST has been plotted on the map, as a general agreement on their definition exists. For the other intervals contemplated within TEMPUS, a generally accepted definition of the precise interval still does not exist. We have thus plotted on the maps a section of the time series that correspond to consecutive intervals for cores that have a reliable age model. The data have been compiled in two figures that encompass those intervals between modern and LGM values and those between the LGM and the Eemian. This approach to plotting the data allows a global comparison of the data and will not require re-plotting of the values if the definition of a particular interval is reassessed in the future. These data will be made available for climate modellers to employ in addition to (or in comparison with) the existing CLIMAP data.