Brief description
The purpose of this study was to estimate past sea-surface temperature in the Southeast Indian Ocean. The accumulation of diatom microplankton assemblages in sediments being the proxy for these records of climatic change.A transfer function is a mathematical method that allows quantitative estimations of environmental parameters from past biogenic sedimentation. In this study, a diatom transfer function was established to relate biological species to selected environmental parameter of the ocean (sea-surface temp). These equations were then employed on observed fossil species from core samples to provide quantitative estimates of sea-surface temperature.
Estimates of sea-surface temperature are given here based on the assemblages of diatoms found in core MD88-787. The information provided gives the extrapolated age-scale through the core, and two estimates of feb SST based on different diatom transfer functions (DTF 109/24/6 and DTF 166/34/4).
It is probable that the 'real' SST signal lies between the model estimates. The model diagnostics and lower number of non-analogue events from DTF 109/24/6 over that of results produced by DTF 166/34/4 suggest that SST estimates are closer to those provided by the former model.
Lineage
Maintenance and Update Frequency: notPlanned
Statement: Three sources of data are procured to perform the Diatom Transfer Function (DTF). The first set taken is the ranked diatom species data from core tops (i.e. Antarctic Diatom Database). The second data set is taken from the modern environmental variable to be related. In this study, the present-day reference for sea-surface temperature, the World Ocean Atlas (WOA, 1994) is used. The third dataset comprises the down-core data which contains the ranked data derived from the same diatom species used in the modern database.
The main differences between the DTF 166/34/4 and the DTF 109/24/6 are related to the number of factors involved, the inclusion of the dissolution data and the differences in the species identified. In terms of the models output, the DTF 166/34/4 will provide estimates that can be in error by up to ±4oC through the 0 to 12oC range, most of the abnormalities being referable to the inclusion of the warm-water dissolution samples. The DTF 109/24/6 under-estimates in the 4-8oC range and over-estimates in the 0-4oC range, but in each case the residuals are confined to within ±2oC. The model does not take into account dissolution of the diatoms in the samples, but does increase the number of factors to account for the wider distributional patterns.
It is probable that the 'real' SST signal lies between the model estimates. The model diagnostics and lower number of non-analogue events from DTF 109/24/6 over that of results produced by DTF 166/34/4 suggest that SST estimates are closer to those provided by the former model.
Cooler estimates provided by the DTF have been shown in the reference 109 database to be possibly over-estimated by up to 1.5oC. The answer may simply lie in the fact that the 109 database is too small and restrictive.
Parameters for Antarctic Diatom Database: Diatom species, Leanne K. Armand code, core code, core coordinates (degrees, decimal minutes), total number of specimens recorded in each core, % of old species, % of data used in 24 major species
Parameters for historical sea-surface temperature estimates: Core depth (cm), analyseries extrapolated age scale (ka) at 10cm interval, DTF commonality, SST Feb (oC), estimate of sea-ice February concentration (%), standard error sea-ice February concentration (%), estimate of sea-ice September concentration (%), standard error sea-ice September concentration (%), estimate of sea-ice months / year cover (mths per year), standard error sea ice month/year cover.
Parameters for the raw diatom counts: Core depth (cm), diatom species, diatom subtotal and silica subtotal
The main differences between the DTF 166/34/4 and the DTF 109/24/6 are related to the number of factors involved, the inclusion of the dissolution data and the differences in the species identified. In terms of the models output, the DTF 166/34/4 will provide estimates that can be in error by up to ±4oC through the 0 to 12oC range, most of the abnormalities being referable to the inclusion of the warm-water dissolution samples. The DTF 109/24/6 under-estimates in the 4-8oC range and over-estimates in the 0-4oC range, but in each case the residuals are confined to within ±2oC. The model does not take into account dissolution of the diatoms in the samples, but does increase the number of factors to account for the wider distributional patterns.
It is probable that the 'real' SST signal lies between the model estimates. The model diagnostics and lower number of non-analogue events from DTF 109/24/6 over that of results produced by DTF 166/34/4 suggest that SST estimates are closer to those provided by the former model.
Cooler estimates provided by the DTF have been shown in the reference 109 database to be possibly over-estimated by up to 1.5oC. The answer may simply lie in the fact that the 109 database is too small and restrictive.
Parameters for Antarctic Diatom Database: Diatom species, Leanne K. Armand code, core code, core coordinates (degrees, decimal minutes), total number of specimens recorded in each core, % of old species, % of data used in 24 major species
Parameters for historical sea-surface temperature estimates: Core depth (cm), analyseries extrapolated age scale (ka) at 10cm interval, DTF commonality, SST Feb (oC), estimate of sea-ice February concentration (%), standard error sea-ice February concentration (%), estimate of sea-ice September concentration (%), standard error sea-ice September concentration (%), estimate of sea-ice months / year cover (mths per year), standard error sea ice month/year cover.
Parameters for the raw diatom counts: Core depth (cm), diatom species, diatom subtotal and silica subtotal
Statement: Use of Alfred-Wegner-Institut computer programs PalaeoToolBox and MacTransfer (Sieger, Nov 1996) which incorporates the Imbrie and Kipp (1971) subroutines (Cabfac, Regress and Thread) have been used in the work produced here. These programs modernise and simplify the original Factor Analysis Package for PC's (Schrader 1989).
Notes
CreditCentre des Faibles Radioactivities mixte CNRS CEA (France)
Credit
Institute of Antarctic and Southern Ocean Studies (IASOS)
Institute of Antarctic and Southern Ocean Studies (IASOS)
Credit
Institut Français pour la Recherche et Technologie Polaires (IFRTP)
Institut Français pour la Recherche et Technologie Polaires (IFRTP)
Credit
Terres Australes et Antarctiques Françaises (TAAF)
Terres Australes et Antarctiques Françaises (TAAF)
Credit
Australian Postgraduate Research Award (APRA)
Australian Postgraduate Research Award (APRA)
Purpose
To estimate past sea-surface temperature in the southeast Indian Ocean
To estimate past sea-surface temperature in the southeast Indian Ocean
Issued: 18 09 2007
Data time period: 1988 to 1988
text: westlimit=145; southlimit=-56.50; eastlimit=145.5; northlimit=-56.00
text: uplimit=3030.4; downlimit=3020
Subjects
BIOLOGICAL CLASSIFICATION |
Diatoms |
EARTH SCIENCE |
MICROALGAE |
OCEAN TEMPERATURE |
OCEANS |
PLANTS |
Sea Surface Temperature |
Transfer function |
biota |
core_depth |
diatom_transfer_function_communality |
diatom_transfer_function_sea_surface_temperature |
environment |
extrapolated_age_scale |
latitude |
longitude |
taxonomic_group_count |
taxonomic_name |
User Contributed Tags
Login to tag this record with meaningful keywords to make it easier to discover
Other Information
Antarctic Diatom Database (App 3-2.pdf)
Historical February sea-surface temperature estimates at location of core MD88-787 (App_6-8__787.pdf)
Raw diatom counts through core MD88-787 (App 6-6 787.pdf)
Identifiers
- global : 274d3e20-6715-11dc-88d6-00188b4c0af8
