1. Ice Core Drilling And Age Scale
The Jurassic ice core (JUR) was drilled in the austral summer 2012/2013 CE to a depth of 140 meters in the south-western side of the Antarctic Peninsula (74.33°S, 73.06°W, 1139 m a.s.l.) (Fig. 1). This ice core was drilled using the British Antarctic Survey (BAS) electromechanical, 104 mm diameter drill. An ice chronology was established based on the annual cycles of hydrogen peroxide (H2O2) and the non-sea salt component of sulphate (nssSO42−)51. The ice chronology was also corroborated using major enhancements in the nssSO42− signal caused by two large volcanic eruptions, Krakatoa (1883 CE) and Pinatubo (1991 CE)52, 53. The Jurassic ice core was dated back to 1873 CE, with an estimated dating error for the 1873–2013 CE interval of ± 4 months for each year and with no accumulated error.
2. Sample Preparation And Analyses
Ice core samples were cut using a band-saw with a steel blade. Discrete samples were cut from the inner part of the ice core at annual resolution for diatom analyses (1.2 cm x 5 cm x depth equivalent of each year). Ice samples for diatoms analyses were melted and then meltwater was filtered through polycarbonate membrane filters (pore diameter 1.0 µm). Filters were subsequently scanned in a Scanning Electron Microscope (SEM), following the method presented in Tetzner et al. (2021)54.
Diatom identification and ecological associations were based on published Southern Ocean databases and references therein55, 56, 57, 58. Observations of diatom preservation were based on the visual recognition of characteristic frustule dissolution and degradation features59. Diatom frustules and fragments with a long axis less than 5 µm were excluded from counting and identification. Diatom counts per sample integrated all diatom valves, partially obscured diatom valves, and fragments with diatom ornamentation identified in each sample. Diatoms were first identified to species level where possible. Diatoms that were impossible to identify unequivocally due to image resolution were combined in genera/morphological groups. Among the reasons preventing the classification of diatoms were the occurrence of valves partly obscured by insoluble particles lying on top, fragments with undiagnostic features, and poorly ornamented or indistinctive fragments. The unclassified diatoms were omitted from assemblage composition and ecological associations but were included in the total diatom counts. After processing, diatom counts per sample (n) were represented as their temporal equivalent, diatom abundance (n yr− 1). The diatom abundance parameter includes all diatoms and diatom remains identified on each sample, regardless of their potential source.
The main diatom assemblage composition was established from the identified species and groups with abundances higher than 2.0% of the whole assemblage. Ecological associations were determined for the most abundant species/groups. Based on ecological associations, diatoms were classified into two groups differentiated by their Antarctic endemism. An Antarctic Diatom Index (ADI) was created to explore the occurrence of periods when one diatom group prevailed over the other. The ADI was obtained by normalizing the diatom abundance of each group and then subtracting them.
3. Statistical Analyses
Changepoint analyses were performed on annual diatom abundance time series to identify the time when the root-mean-square level of the power spectral density signal presented the most significant change. Changepoint calculations were conducted using a ten-year window with nine overlapping data points60. Changepoint calculations were conducted using the “findchangepts” function in Matlab 2021 software.
Trends in the diatom abundance record were identified using the empirical mode decomposition (EMD) approach61, 62, 63, 64. EMD is an algorithm which decomposes a time series into a finite number of intrinsic mode functions and a nonlinear trend component. The statistical significance of the nonlinear trend is estimated using the Monte Carlo approach63.
4. Climate Analyses
Monthly reanalysis fields from the fifth generation of the European Center for Medium-Range Weather Forecasts (ECMWF), ERA565, were used to obtain spatial correlations between the diatom abundance and wind speed (10 m wind speed). ERA5 datasets provide hourly data available at 0.25-degree resolution (~ 31 km) since 1979 CE.
Method References
51. Emanuelsson, B. D., Thomas, E. R., Tetzner, D. R., Humby, J. D., & Vladimirova, D. O. Ice core chronologies from the Antarctic Peninsula: the Palmer, Jurassic, and Rendezvous Age-scales. Geosciences, 12(2), 87. (2022).
52. Cole‐Dai, J., Mosley‐Thompson, E., & Thompson, L. G. Annually resolved southern hemisphere volcanic history from two Antarctic ice cores. Journal of Geophysical Research: Atmospheres, 102(D14), 16761-16771. (1997).
53. Tetzner, D. R., Thomas, E. R., Allen, C. S., & Piermattei, A. Evidence of recent active volcanism in the Balleny Islands (Antarctica) from ice core records. Journal of Geophysical Research: Atmospheres, 126(23), e2021JD035095. (2021).
54. Tetzner, D., Thomas, E. R., Allen, C. S., & Wolff, E. W. A refined method to analyze insoluble particulate matter in ice cores, and its application to diatom sampling in the Antarctic Peninsula. Frontiers in Earth Science, 9, 617043. (2021).
55. Armand, L. K., Crosta, X., Romero, O., and Pichon, J. J. The biogeography of major diatom taxa in Southern Ocean sediments: 1. Sea ice related species. Palaeogeogr. Palaeoclimatol. Palaeoecol. 223 (1–2). (2005).
56. Hasle, G. R., & Syvertsen, E. E. Identifying Marine Phytoplankton. Academic Press, London. (1997).
57. Cefarelli, A. O. et al. Diversity of the diatom genus Fragilariopsis in the Argentine Sea and Antarctic waters: morphology, distribution and abundance. Polar Biol.33 (11), 1463–1484. (2010).
58. Zielinski, U., and Gersonde, R. Diatom distribution in Southern Ocean surface sediments (Atlantic sector): implications for paleoenvironmental reconstructions. Palaeogeogr. Palaeoclimatol. Palaeoecol. 129 (3–4), 213–250. (1997).
59. Warnock, J. P., and Scherer, R.P. A revised method for determining the absolute abundance of diatoms. J. Paleolimnol. 53(1),157–163. (2015).
60. Killick, R., Fearnhead, P., & Eckley, I. A. Optimal detection of changepoints with a linear computational cost. Journal of the American Statistical Association, 107(500), 1590-1598. (2012).
61. Huang, N. E. et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis. Proceedings of the Royal Society of London. 454(1971), 903-995. (1998).
62. Huang, N. E., & Wu, Z. A review on Hilbert‐Huang transform: Method and its applications to geophysical studies. Reviews of geophysics, 46(2). (2008).
63. Franzke, C. Multi-scale analysis of teleconnection indices: climate noise and nonlinear trend analysis. Nonlinear Processes in Geophysics, 16(1), 65-76. (2009).
64. Thomas, E. R., Dennis, P. F., Bracegirdle, T. J., & Franzke, C. Ice core evidence for significant 100‐year regional warming on the Antarctic Peninsula. Geophysical Research Letters, 36(20). (2009).
65. Hersbach, H., Dee, D. ERA5 reanalysis is in production. ECMWF Newsl,147, 7. (2016).