Quaternary surface water temperature estimations: New diatom transfer functions for the Southern Ocean

https://doi.org/10.1016/j.palaeo.2014.08.008Get rights and content

Highlights

  • Reconstruction of Quaternary summer sea surface temperatures in the Southern Ocean

  • Integration of diatoms from 336 sites of the Pacific, Atlantic and Indian sectors

  • Comprehensive statistical testing of the surface sediment reference data set

  • Four statistical approaches to reconstruct Southern Ocean summer temperatures

  • Testing of the transfer functions on two down-core records

Abstract

Here we present an improvement of paleotemperature reconstructions for the Southern Ocean by combining new diatom data from the Pacific sector with published Atlantic and Indian sector reference data. The statistical analysis of 336 surface sediment samples recovered from a wide area of Southern Ocean environments defines a supra-regional reference data set for quantitative summer sea surface temperature (SSST) estimations. In situ temperature measurements covering the time span from approx. 1900 to 1991 were used as reference instead of more recent time series of satellite-derived data, possibly biased by ocean warming.

Different transfer function (TF) models for the Imbrie and Kipp Method (IKM), the Modern Analog Technique (MAT), Weighted Averaging (WA), and Weighted Averaging Partial Least Squares (WAPLS) were tested. Best performance for IKM was obtained using the D336 set with 29 diatom taxa and three factors, resulting in root mean square error of prediction (RMSEP) of 0.833 °C for SSST. MAT estimates were best with six analogs resulting in the lowest RMSEP of 0.812 °C. WAPLS applied to D336 resulted in a RMSEP of 0.782 °C. WA performed less well, expressed by a RMSEP of 0.974 °C. Furthermore, two subsets for the Atlantic (D151) and the Pacific sectors (D107) were applied with IKM to test for the advantages of localized TFs. IKM-D151 and IKM-D107 performed comparably good as IKM-D336, with RMSEP of 0.71 °C and 0.68 °C, respectively.

Application of the augmented reference data sets on two Pleistocene sediment records from the Atlantic (PS1768-8) and Pacific (PS58/271-1) sectors led to the best performance of IKM with D336, expressed by high overall communalities (> 0.75) and fewer (PS1768-8) to no (PS58/271-1) no-analogs, compared to the regional data sets, proving IKM-D336 to deal better with higher assemblage variability. SSST estimates for both cores exhibit similar glacial/interglacial patterns for all four applied D336-based TF methods, with the best concordance between IKM and WAPLS.

Introduction

The estimation of paleoenvironmental variables from micropaleontological data sets has become an important tool for paleoceanographic reconstructions since the development of the Imbrie and Kipp transfer function method (IKM; Imbrie and Kipp, 1971) and the Modern Analog Technique (MAT; Hutson, 1980). After first comprehensive and successful use of the IKM on various microfossil groups in the Climate: Long-Range Investigation Mapping and Prediction (CLIMAP) project for the reconstruction of the last glacial maximum (LGM) and the last interglacial optimum (Climate: Long-Range Investigation, Mapping and Prediction (CLIMAP), 1976, Climate: Long-Range Investigation, Mapping and Prediction (CLIMAP), 1981, Climate: Long-Range Investigation, Mapping and Prediction (CLIMAP), 1984), and the successful application of the MAT proposed by Hutson (1980), Sea Surface Temperature (SST) estimations derived from transfer functions (TFs) are broadly accepted and applied in paleoceanography (e.g. Crosta et al., 1998, Crosta et al., 2004, Zielinski et al., 1998, Gersonde and Zielinski, 2000, Gersonde et al., 2005). Further improvement of TFs to deal with non-linear species responses to environmental gradients led to the development of methods based on Weighted Averaging (WA; Birks et al., 1990) and Weighted Averaging Partial Least Squares (WAPLS; ter Braak and Juggins, 1993), successfully applied in diatom-based SST reconstructions (e.g. Birks and Koc, 2002). Potentials and limitations concerning the application of TFs are summarized e.g. in Birks (2003), Telford and Birks (2005) and Juggins (2013). A method for assessing the statistical significance of quantitative reconstructions inferred from microfossil assemblages is proposed by Telford and Birks (2011).

In the Southern Ocean, diatom assemblages are useful tools for the reconstruction of past SSTs because diatoms are widely distributed, they are phototrophic algae living in the euphotic zone, thus representing environmental conditions of the upper water column, and their biogeographic distribution patterns are closely related to surface water summer temperature (e.g. Zielinski and Gersonde, 1997, Armand et al., 2005, Crosta et al., 2005, Romero et al., 2005, Esper et al., 2010). However, in order to reconstruct Pleistocene summer SSTs based on diatom assemblages, several limitations have to be taken into account. This includes selective dissolution of less robustly silicified diatom valves resulting in an overall dominance of a few robust species in the surface and the down-core sediment record (Zielinski et al., 1998, Esper and Gersonde, 2014). Another limitation are so-called non-analog conditions, where fossil assemblages are composed by taxa at numbers that exceed the abundance in the surface sediment reference data set or in case of the prominent occurrence of extinct species. A typical non-analog situation in Southern Ocean diatom assemblages is related to the abundant occurrence of Eucampia antarctica in glacial sediments, leading to the exclusion of such taxa from reference data sets (Zielinski et al., 1998, Esper and Gersonde, 2014). Nevertheless, the previous studies on quantitative SSST reconstructions based on diatoms show a down-core pattern congruent with radiolarian- and foraminifera-derived SSSTs in accordance with ice core temperature variation for the last 150,000 yr (e.g. Gersonde et al., 2004), proving a close diatom assemblage/SSST-relationship for at least this time span.

The previous diatom TFs for the Southern Ocean comprised Atlantic, Indian and a few Pacific sector surface samples (Crosta et al., 1998, Zielinski et al., 1998, Gersonde et al., 2005), calling for a more comprehensive Pacific sector extension of diatom sample collections to allow for reliable quantitative summer SST reconstructions in this largest Southern Ocean sector (Gersonde et al., 2005). The new reference data set presented herein comprises 336 samples with 29 diatom taxa from the Pacific, Atlantic, and western Indian sectors of the Southern Ocean (Fig. 1, Table 1) and expands the reference data set D93-29 of Zielinski et al. (1998). The combination of new diatom data from the Pacific and Atlantic sectors with published data allows for the separation of regional sub-sets for the Atlantic (151 samples) and the Pacific sector (107 samples). A further aim of this study is to compare corresponding sea surface temperatures derived from different hydrographic collections to identify reference SSTs with the best correlation with the sediment-derived diatom assemblages. Additionally, we elaborate, how statistical techniques are applied to optimize the estimation capacity and reliability of four different TF approaches for their application in circum-Antarctic environments. Interpolation techniques based on linear (IKM) or unimodal (WA, WAPLS) species-environment response models will be compared with MAT, a method directly relating modern reference data with down-core diatom assemblages. Finally, we apply the four transfer functions to two Pleistocene sediment cores from the Atlantic (PS1768-8) and eastern Pacific (PS58/271-1) sectors of the Southern Ocean (Fig. 1) to document the past 150 kyr (kyr = 1000 yr) summer SST variability in the Permanent Open Ocean Zone of the Southern Ocean with emphasis on a performance comparison between regional sub-sets and the full reference data set.

Section snippets

Samples, preparation and counting

To generate a new diatom calibration set for summer sea surface temperature (SSST) reconstruction the diatom assemblage data from 467 surface samples recovered in the Atlantic, the western Indian, and the Pacific sectors of the Southern Ocean were examined and tested for their applicability to TF-based SSST estimates. This comprises data sets obtained from 230 sites in the Atlantic sector (Zielinski and Gersonde, 1997), 29 sites in the Atlantic and western Indian sectors (Zielinski et al., 1998

Establishment of a diatom reference data set for TF-based temperature reconstruction

To establish the new diatom reference data set for the SO, the diatom assemblage composition preserved in a total of 467 surface sediment samples was reviewed. As a first step, the number of taxa/taxa groups, originally amounting to 85, was reduced to 35 (Table 1), following general rules proposed by Imbrie and Kipp (1971) and findings concerning the environmental significance of Antarctic diatom taxa established by Zielinski et al. (1998). Thus, taxa occurring at low abundance (generally below

Conclusions

The development of the new reference data set D336/28, including regional subsets D151/29 (Atlantic sector) and D107/27 (Pacific sector), and statistical tests for several pre-settings of transfer functions based on IKM, MAT, WA, and WAPLS led to reliable diatom transfer functions for paleo-SSST reconstruction in the Southern Ocean. Limitations for selecting a valuable reference data set are (1) parameters which alter the diatom assemblage in surface sediment samples (reworking, selective

Acknowledgments

We acknowledge the technical assistance of U. Bock and R. Cordelair. Finally we thank the reviewers for the constructive reviews and comments on the manuscript. The research has received funding from the European Union's Seventh Framework Programme (FP7/2007–2013) under grant agreement no. 243908, ‘Past4Future’ Climate change — Learning from the past climate, the AWI program PACES (“Polar Regions and Coasts in the Changing Earth System”) and MARUM (“The Ocean in the Earth System”).

References (66)

  • M. Kucera et al.

    Reconstruction of sea-surface temperatures from assemblages of planktonic foraminifera: multi-technique approach based on geographically constrained calibration data sets and its application to glacial Atlantic and Pacific Oceans

    Quat. Sci. Rev.

    (2005)
  • J. Le

    Palaeotemperature estimation methods: sensitivity test on two western equatorial Pacific cores

    Quat. Sci. Rev.

    (1992)
  • J. Le et al.

    Reconstructing paleoenvironment by transfer function: model evaluation with simulated data

    Mar. Micropaleontol.

    (1994)
  • A. Leventer

    Modern distribution of diatoms in sediments from the George V Coast, Antarctica

    Mar. Micropaleontol.

    (1992)
  • B.A. Malmgren et al.

    Assessment of quantitative techniques in paleobiogeography

    Mar. Micropaleontol.

    (1982)
  • A.H. Orsi et al.

    On the meridional extend and fronts of the Antarctic Circumpolar Current

    Deep-Sea Res. I

    (1995)
  • O. Romero et al.

    The biogeography of major diatom taxa in Southern Ocean sediments: 3. Tropical/Subtropical species

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (2005)
  • R.J. Telford et al.

    The secret assumption of transfer functions: problems with spatial autocorrelation in evaluating model performance

    Quat. Sci. Rev.

    (2005)
  • R.J. Telford et al.

    A novel method for assessing the statistical significance of quantitative reconstructions inferred from biotic assemblages

    Quat. Sci. Rev.

    (2011)
  • C.J.F. ter Braak

    Non-linear methods for multivariate statistical calibration and their use in palaeoecology: a comparison of inverse (k-nearest neighbours, partial least squares and weighted averaging partial least squares) and classical approaches

    Chemom. Intell. Lab. Syst.

    (1995)
  • U. Zielinski et al.

    Diatom distribution in Southern Ocean surface sediments (Atlantic sector): implications for paleoenvironmental reconstructions

    Palaeogeogr. Palaeoclimatol. Palaeoecol.

    (1997)
  • J.I. Antonov et al.

    World Ocean Atlas 2009, Volume 2: Salinity

  • H.J.B. Birks

    Quantitative Palaeoenvironmental Reconstructions

  • H.J.B. Birks

    Quantitative Palaeoenvironmental Reconstructions from Holocene Biological Data

  • C.J.A. Birks et al.

    A high-resolution diatom record of late-Quaternary sea-surface temperatures and oceanographic conditions from the Norwegian Sea

    Boreas

    (2002)
  • H.J.B. Birks et al.

    Diatoms and pH reconstruction

    Phil. Trans. R. Soc. London B

    (1990)
  • C.W. Böning et al.

    The response of the Antarctic Circumpolar Current to recent climate change

    Nat. Geosci.

    (2008)
  • R.B. Cattell

    The scree test for the number of factors

    Multivar. Behav. Res.

    (1966)
  • Climate: Long-Range Investigation, Mapping and Prediction (CLIMAP)

    The surface of the ice-age Earth

    Science

    (1976)
  • Climate: Long-Range Investigation, Mapping and Prediction (CLIMAP)

    Seasonal reconstructions of the Earth's surface at the last glacial maximum

    Geol. Soc. Am. Map Chart Ser.

    (1981)
  • Climate: Long-Range Investigation, Mapping and Prediction (CLIMAP)

    The last interglacial ocean

    Quat. Res.

    (1984)
  • X. Crosta et al.

    Application of modern analog technique to marine Antarctic diatoms: reconstruction of maximum sea-ice extent at the Last Glacial Maximum

    Paleoceanography

    (1998)
  • W.L. Cunningham et al.

    Diatom assemblages in surface sediments of the Ross Sea: relationship to present oceanographic conditions

    Antarct. Sci.

    (1998)
  • Cited by (45)

    • How does the Southern Ocean palaeoenvironment during Marine Isotope Stage 5e compare to the modern?

      2022, Marine Micropaleontology
      Citation Excerpt :

      Diatom relative abundances for each sediment core are reported for species or groups with well-constrained present-day ecologies/habitats (Table 2). Actinocyclus actinochilus is a cold-water species (Table 2) generally found in low abundances (<3%) in SO seafloor surface sediment samples within the maximum WSIE (Armand et al., 2005; Esper and Gersonde, 2014b). Increasing relative abundances of this species in southern high latitude sediments suggest colder SSTs and more severe sea-ice cover.

    • Southern Ocean sea surface temperature synthesis: Part 1. Evaluation of temperature proxies at glacial-interglacial time scales

      2021, Quaternary Science Reviews
      Citation Excerpt :

      Diatom-rich ‘marine snow’ sinking rates may also depend on species composition (Laurenceau-Cornec et al., 2015); this could significantly skew sedimentary assemblages. Given the slow sinking rate of individual diatoms, yet the relatively low RMSEP for this proxy (e.g., Esper and Gersonde, 2014), we suggest that sedimentary diatoms mostly reach the ocean floor in aggregates or faecal pellets. However, we are not confident in the validity of this assumption, and even these aggregates could still undergo significant advection in some regions.

    View all citing articles on Scopus
    View full text