Preprint: The Influence of Southern Hemisphere Climate Variability on Water Isotopes from the Roosevelt Island ice core, Antarctica

• Main Author: Emanuelsson, Daniel
• Format: Text
• Language: unknown
• Published: figshare 2018
• Subjects: 40102 Atmospheric Dynamics; FOS Earth and related environmental sciences; Climate Science; 40105 Climatology excl. Climate Change Processes; Paleoclimatology; Pacific; Roosevelt Island; Ross Ice Shelf; Antarc*; Antarctica; ice core; Ice Shelf; Sea ice
• Online Access: https://dx.doi.org/10.6084/m9.figshare.7072664; https://figshare.com/articles/Preprint_The_Influence_of_Southern_Hemisphere_Climate_Variability_on_Water_Isotopes_from_the_Roosevelt_Island_ice_core_Antarctica/7072664

In this study, the three leading patterns of Southern Hemisphere (SH) atmospheric circulation (z500), sea ice concentration (SIC) and near-surface air temperatures (SAT) variability’s influence on the Roosevelt Island Climate Evolution (RICE) ice-core deuterium isotopes (δD) are investigated using empirical orthogonal function (EOF) analysis. Roosevelt Island is located near the coast at the northeastern margin of the Ross Ice Shelf, Antarctica. We provide a methodology that explains the z500, SAT and SIC correlation strength with RICE δD at any given point in the SH. Therefore, the method encompasses an explanation for why the RICE δD relationship with local SAT is not significant. We show significant relationships between the first leading patterns of variability and between the second z500 [the first Pacific–South American pattern (PSA1)], SIC and the third SAT EOF patterns. We demonstrate that δD is correlated with all the 1–3 leading patterns of SH variability except with z500 EOF3 (PSA2), SIC EOF3 and SAT EOF2. Combined the lack of significant influence on δD from some leading patterns and the relation between the EOF patterns with significant influence on δD explains the δD SH correlation patterns. For example, for SAT, the lack of influence form EOF2 and the in-phase relation between SAT EOF1 and SAT EOF3, which strengthen and cancel the correlation over extensive SH regions, explains the δD-SAT correlation patterns and why the RICE δD correlation with local SAT does not exceed r = 0.36 (p < 0.1, 1979–2011).