Abrupt climate changes for Iceland during the last millennium: Evidence from high resolution sea ice reconstructions

https://doi.org/10.1016/j.epsl.2008.03.017Get rights and content

Abstract

A high resolution account of Icelandic sea ice over the last millennium has been constructed using a novel proxy based on the presence in sediments of a biomarker (IP25) produced by sea ice algae. Comparison with historical sea ice records demonstrates a strong correlation between documented sea ice occurrences and the IP25 proxy. An excellent agreement is also observed between the IP25 record and a diatom-based sea surface temperature reconstruction obtained from the same core and the Crowley and Lowery Northern Hemisphere temperature reconstruction. Using this approach, we provide new historical sea ice data for periods where records are scarce or absent and evidence for abrupt changes to sea ice and/or climate conditions around Iceland during the Little Ice Age.

Introduction

Given the current debate regarding climate change on Earth and, in particular, the relative contributions of natural processes and anthropogenic inputs, it is crucial to obtain a clear and detailed account of past climatic variations and the factors controlling these (Jones et al., 2001). Polar sea ice, by its influence on the heat exchanges between the oceans and the atmosphere and its contributions to numerous oceanic processes (e.g. thermo-haline circulation) is a key component of the Earth's climate system (Thomas and Dieckmann, 2003). Therefore, improving our knowledge of historical sea ice fluctuations at a high spatial and temporal resolution will help to refine future climate change models and improve predictions. Very few documentary records for sea ice exist which pre-date the instrument era and these often include unreliable data (Bergthórsson, 1969, Ogilvie and Jónsson, 2001). Here, we report a detailed analysis of a sediment core (MD99-2275) collected from the North Icelandic Shelf (Fig. 1). This area is under the strong influence of three surface currents (Fig. 1). The warm and high salinity Irminger current is a branch of the North Atlantic drift, travelling along the western and the north western coasts of Iceland, while the East Greenland and the East Icelandic currents bring cold and low salinity polar waters to the region. Any change in the relative strengths of these currents will influence the position of the oceanic Polar Front, and this is likely to be archived in the sediment record (Knudsen et al., 2004).

In this study, we have used the recently established sea ice proxy, IP25 (Belt et al., 2007), which is based on the preservation in marine sediments of a unique chemical fossil produced by sea ice algae (Fig. 2), to obtain an uninterrupted, high resolution (ca. 2–5 yr) record of sea ice occurrences for the last millennium. The very high sedimentation rates associated with the core location, together with well documented occurrences of volcanic tephras (Eiriksson et al., 2004, Knudsen and Eiriksson, 2002, Larsen et al., 2002, Rousse et al., 2006) have enabled us to perform this study at an unprecedented sub-decadal resolution and to make comparisons with historical data documenting past sea ice extending back to the early days of Icelandic colonization (ca. 1080 BP). We demonstrate strong correlations between documented sea ice occurrences and the IP25 proxy (Bergthórsson, 1969, Ogilvie and Jónsson, 2001) and reveal new sea ice data for periods where historical sources are scarce or absent. We have also compared our IP25 data with diatom-based sea surface temperature reconstructions (Jiang et al., 2005, Eiriksson et al., 2006) to confirm that the Icelandic climate was relatively mild, and that little sea ice occurred in the region from 800 to 1300 AD, corresponding to the end of the Warm Mediaeval Period (MWP). In contrast, both reconstructed sea ice and sea surface temperatures show that the climatic conditions around northern Iceland worsened during the second part of the millennium with cooler sea surface temperatures and larger amounts of sea ice. More detailed correlations exist throughout the record, and also provide evidence for a succession of abrupt climate changes in Iceland during the latter part of the record, corresponding to the Little Ice Age (LIA, 1300–1900 AD). Finally, our sea ice record also shows strong correlations with hemispheric-scale temperature reconstructions (Crowley, 2000, Mann et al., 1998), indicating that climatic conditions over Iceland were representative for at least the last millennium. This case study demonstrates that IP25 is a reliable proxy for historical sea ice reconstructions and could become an invaluable tool for high or ultra-high resolution studies of the Earth's climate system.

Section snippets

Sediment samples

The core MD99-2275 (66 33.06'N, 17 41, 59'W; 410 m water depth) was collected during the R/V Marion Dufresne IMAGES V cruise in 1999. The age model of the entire core was determined using a combination of tephra marker horizons and thirty five radiocarbon dates (Eiriksson et al., 2004, Knudsen and Eiriksson, 2002, Larsen et al., 2002, Rousse et al., 2006). This age model was further constrained using 6 tephra layers corresponding to the time period examined in the present study according to the

Results and discussion

Palaeoclimate scientists continue to emphasise the importance of data derived from so-called proxy methods for climate reconstruction and that such data should have both high temporal and spatial resolution if it is to be valuable for both historical determinations and future climate prediction models (Mann et al., 1998, Jones et al., 1998, Jones et al., 2001). Such studies routinely rely on a multi-proxy approach since direct measures of climate conditions are either scarce, absent or

Conclusion

This first application of a novel sea ice proxy has involved a comparison between the abundances of a sea ice derived biomarker found in an Icelandic sediment core with historical sea ice records, diatom-based sea surface temperatures and mean northern hemisphere temperatures. For the last millennium, we demonstrate a significant set of correlations between the abundance of the IP25 biomarker and at least one (and often two or all three) of these other climatic measures. As such, we have been

Acknowledgements

This work was supported by the UK Natural Environment Research Council (NE/D013216/1; NE/E00752X/1). This is a contribution to the European Union 5th Framework project PACLIVA (Contract No. ECK2-CT-2002-00143). This is LSCE contribution number 2435.

Cited by (194)

  • Predicting the future of coastal marine ecosystems in the rapidly changing Arctic: The potential of palaeoenvironmental records

    2022, Anthropocene
    Citation Excerpt :

    The resulting inference model is subsequently applied to down-core fossil assemblages, providing quantitative reconstructions of environmental parameters back in time (e.g. Koç et al., 1993; de Vernal et al., 2001; Ledu et al., 2010; Miettinen et al., 2015; Sha et al., 2016; Krawczyk et al., 2017; Falardeau et al., 2018) (Fig. 2). A rapid rise of sea-ice reconstructions based on HBI biomarkers has followed the introduction of IP25 (Belt et al., 2007), to date the most well-known and most widely used of the HBIs (e.g. Massé et al., 2008; Vare et al., 2010; Müller et al., 2012; Cabedo-Sanz et al., 2013; Méheust et al., 2016; Detlef et al., 2018), which is produced by certain sea-ice dwelling diatom species (Brown et al., 2014; Limoges et al., 2018a). For a detailed recent review on their development, strengths and constraints, see Belt (2018), who also discusses ratio-based semi-quantitative sea-ice indices using HBIs and phytoplankton markers (e.g. PIP25, PIIIP25).

  • Sea ice in the northern North Atlantic through the Holocene: Evidence from ice cores and marine sediment records

    2021, Quaternary Science Reviews
    Citation Excerpt :

    The quartz percentage is generally higher during the last millennium (Moros et al., 2006, Fig. 3f) and the IP25 values reach their Holocene maximum (Cabedo-Sanz et al., 2016, Fig. 3f). The marked increase in IP25 from ca. 0.7 ka until the 20th century is consistent with an earlier IP25 study (Massé et al., 2008, not shown) from another core from the North Iceland shelf. Both IP25 records also record relatively low values in the early part of the last millennium, suggesting reduced sea ice at this time.

View all citing articles on Scopus
View full text