ALBERT

Description: The causes for rising temperatures along the Antarctic Peninsula during the late Holocene have been debated, particularly in light of instrumental records of warming over the past decades (Russell and McGregor, 2010, doi:10.1007/s10584-009-9673-4). Suggested mechanisms range from upwelling of warm deep waters onto the continental shelf in response to variations in the westerly winds (Bentley et al., 2009, doi:10.1177/0959683608096603), to an influence of El Niño-Southern Oscillation on sea surface temperatures (Shevenell et al., 2011, doi:10.1038/nature09751). Here, we present a record of Holocene glacial ice discharge, derived from the oxygen isotope composition of marine diatoms from Palmer Deep along the west Antarctic Peninsula continental margin. We assess atmospheric versus oceanic influences on glacial discharge at this location, using analyses of diatom geochemistry to reconstruct atmospherically forced glacial ice discharge and diatom assemblage (Taylor and Sjunneskog, 2002, doi:10.1029/2000PA000564) ecology to investigate the oceanic environment. We show that two processes of atmospheric forcing-an increasing occurrence of La Niña events (Makou et al., 2010, doi:10.1130/G30366.1) and rising levels of summer insolation-had a stronger influence during the late Holocene than oceanic processes driven by southern westerly winds and upwelling of upper Circumpolar Deepwater. Given that the evolution of El Niño-Southern Oscillation under global warming is uncertain (Yeh et al., 2009, doi:10.1038/nature08316), its future impacts on the climatically sensitive system of the Antarctic Peninsula Ice Sheet remain to be established.

Description: Lake sediment samples were taken in April 2013 from the ice by drilling through lake ice and recovering an undisturbed core using a HON-Kajak sediment corer. Samples were analysed for pigments (University of Nottingham), carbon isotopes and C/N ratios (BGS, Keyworth), lipid biomarkers (Newcastle University) and compound-specific carbon isotopes (CUG, Wuhan). The purpose of the analyses was to develop an environmental reconstruction of carbon cycling for an upland lake (named Disko 2) to encompass the Little Ice Age to recent warming climate periods. Analyses were completed as part of Mark A. Stevenson's PhD research while based at the University of Nottingham, UK (Stevenson, 2017, http://eprints.nottingham.ac.uk/46579). ²¹⁰Pb, ²²⁶Ra, ¹³⁷Cs and ²⁴¹Am concentrations were measured by direct gamma assay in the Environmental Radiometric Facility at University College London (Dr Handong Yang), using an ORTEC HPGe GWL series well-type coaxial low background intrinsic germanium detector. Radiometric dating techniques follow Appleby et al, 1986 (doi: 10.1007/BF00026640), Appleby et al, 1992 (doi:10.1016/0168-583X(92)95328-O) and Appleby, 2001 (doi:10.1007/0-306-47669-X_9) with core extrapolation and linear interpolation used to derive an age depth model to the base of the core. The pigment β-carotene was analysed on an Agilent 1200 series high-performance liquid chromatography (HPLC) using separation conditions outlined in McGowan et al., 2012 (doi:10.1111/j.1365-2427.2011.02689.x). Bulk δ¹³C and C~org~/N ratios were analysed on acidified samples using a Costech ECS4010 elemental analyser (EA) coupled to a VG Triple Trap and a VG Optima dual-inlet mass spectrometer. Key lipid biomarkers (n-alkanes, n-alkanoic acids (as fatty acid methyl esters (FAMEs), n-alkanols and sterols) were analysed using an Agilent 7890A GC coupled to a 5975C MS according to Pearson et al., 2007 (doi:10.1016/j.orggeochem.2007.02.007) and are expressed as ratios, relative to the total of each compound class. Specific ratios were also calculated for CPI 2 n-alkanes (Marzi et al., 1993; doi:10.1016/0146-6380(93)90016-5), terrestrial aquatic ratio (TAR) for n-alkanes (Bourbonniere and Meyers, 1996; doi:10.1007/s002540050074), index of waxy n-alkanes to total hydrocarbons (PWAX) (Zheng et al., 2007; doi:10.1016/j.orggeochem.2007.06.012) and carbon preference index (CPI) for n-alkanoic acids (Matsuda and Koyama, 1977) (doi:10.1016/0016-7037(77)90214-9). Compound-specific δ¹³C on C~28:0~ fatty acid methyl ester (FAME) was analysed using a Thermo Finnigan Trace GC coupled to a Thermo Finnigan Delta Plus XP isotope ratio mass spectrometer using a combustion interface (GC-C-IRMS) according to conditions in Huang et al. (2018; doi:10.1038/s41467-018-03804-w). Acknowledgements: Mark Stevenson gratefully acknowledges the receipt of a NERC/ESRC studentship (ES/J500100/1). We acknowledge grants IP-1393-1113 & IP-1516-1114 from the NERC Isotope Geosciences laboratory (NIGL) for the analysis of δ¹³C~org~ & C/N ratios on sediment, soil and plant samples. Lipid and water chemistry analyses were funded by the Freshwater Biological Association's 2015 Gilson Le Cren Memorial Award to Mark Stevenson. We thank Teresa Needham, Christopher Kendrick, Julie Swales, Ian Conway, Graham Morris, Bernard Bowler, Paul Donohoe, Qingwei Song and Jiantao Xue for technical support. We acknowledge the support of Handong Yang for radiometric dating. Financial support for fieldwork was awarded via the INTERACT transnational access scheme (grant agreement No 262693) under the European Community's Seventh Framework Programme and UK RI NERC grant NE/K000276/1. Logistical support is acknowledged from University of Copenhagen Arktisk Station including Ole Stecher, Kjeld Mølgaard and Erik Wille.