Drivers of late Holocene ice core chemistry in Dronning Maud Land: the context for the ISOL-ICE project

Within the framework of the Isotopic Constraints on Past Ozone Layer in Polar Ice (ISOL-ICE) project, we present initial ice core results from the new ISOL-ICE ice core covering the last millennium from high-elevation Dronning Maud Land (DML) and discuss the implications for interpreting the stable isotopic composition of nitrogen in ice core nitrate (δ15N(NO)) as a surface ultra-violet radiation (UV) and total column ozone (TCO) proxy. In the quest to derive TCO using δ15N(NO), an understanding of past snow accumulation changes, as well as aerosol source regions and present-day drivers of their variability, is required. We therefore report here the ice core age–depth model, the snow accumulation and ice chemistry records, and correlation analysis of these records with climate variables over the observational era (1979–2016). The ISOL-ICE ice core covers the last 1349 years from 668 to 2017 CE ± 3 years, extending previous ice core records from the region by 2 decades towards the present and shows excellent reproducibility with those records. The extended ISOL-ICE record of last 2 decades showed a continuation of the methane sulfonate (MSA−) increase from ∼ 1800 to present while there were less frequent large deposition events of sea salts relative to the last millennium. While our chemical data do not allow us to distinguish the ultimate (sea ice or the open ocean) source of sea salt aerosols in DML winter aerosol, our correlation analysis clearly suggests that it is mainly the variability in atmospheric transport and not the sea ice extent that explains the interannual variability in sea salt concentrations in DML. Correlation of the snow accumulation record with climate variables over the observational era showed that precipitation at ISOL-ICE is predominately derived from the South Atlantic with onshore winds delivering marine air masses to the site. The snow accumulation rate was stable over the last millennium with no notable trends over the last 2 decades relative to the last millennium. Interannual variability in the accumulation record, ranging between 2 and 20 cm a−1 (w.e.), would influence the ice core δ15N(NO) record. The mean snow accumulation rate of 6.5±2.4 cm a−1 (w.e.) falls within the range suitable for reconstructing surface mass balance from ice core δ15N(NO), highlighting that the ISOL-ICE ice core δ15N(NO) can be used to reconstruct either the surface mass balance or surface UV if the ice core δ15N(NO) is corrected for the snow accumulation influence, thereby leaving the UV imprint in the δ15N(NO) ice core record to quantify natural ozone variability.

Details