PP23B-1398:
A 100-year Reconstruction of Regional Sea Ice Extent in the Ross and Amundsen-Bellingshausen Seas as Derived from the RICE Ice Core, Coastal West Antarctica

Tuesday, 16 December 2014
Daniel B. Emanuelsson1,2, Nancy A.N. Bertler2,3, W Troy Baisden2 and Elizabeth D. Keller2, (1)Victoria University of Wellington, Wellington, New Zealand, (2)GNS Science-Institute of Geological and Nuclear Sciences Ltd, Lower Hutt, New Zealand, (3)Victoria University of Wellington, Antarctic Research Centre, Wellington, New Zealand
Abstract:
Antarctic sea ice increased over the past decades. This increase is the result of an increase in the Ross Sea (RS) and along the coast of East Antarctica, whereas the Amundsen-Bellingshausen Seas (ABS) and the Antarctic Peninsula has seen a general decline. Several mechanisms have been suggested as drivers for the regional, complex sea ice pattern, which include changes in ocean currents, wind pattern, as well as ocean and atmospheric temperature.

As part of the Roosevelt Island Climate Evolution (RICE) project, a 763 m deep ice core was retrieved from Roosevelt Island (RI; W161° 21’, S79°41’, 560 m a.s.l.), West Antarctica. The new record provides a unique opportunity to investigate mechanism driving sea ice variability in the RS and ABS sectors. Here we present the water stable isotope record (δD) from the upper part of the RICE core 0-40 m, spanning the time period from 1894 to 2011 (Fig. 1a).

Annual δD are correlated with Sea Ice Concentration (SIC). A significant negative (r= -0.45, p≤ 0.05) correlation was found between annual δD and SIC in the eastern RS sector (boxed region in Fig. 1b) for the following months NDJFMA (austral summer and fall). During NDJFMA, RI receives local moisture input from the RS, while during the rest of the year a large extent of this local moisture source area will be covered with sea ice with the exception of the RS Polynya. Concurrently, we observe positive δD and SIC correlations in the ABS, showing a dipole pattern with the eastern RS. For this reason, we suggest that the RICE δD might be used as a proxy for past SIC for the RS and ABS region.

There is no overall trend in δD over 100 years (r= -0.08 ‰ dec-1, p= 0.81, 1894-2011). However, we observe a strong increase from 2000-2011 of 17.7 ‰ dec-1(p≤ 0.1), yet the recent δD values and trend of the last decade are not unprecedented (Fig. 1a). We investigate changes in sea surface temperature, atmospheric temperature, inferred surface ocean currents and atmospheric wind fields to determine the driving mechanisms over the observational period (1979-2011).

Figure 1. a) Annual δD RI record 1894-2011 (black dots) and δD decadal running mean (blue line). b) Correlation plot between δD and six-months (NDJFMA) seasonal SIC means between 1980- 2011. Red star indicates location of RI and white contours shows areas where the correlation is significant to ≥95% confidence level.