A glaciochemical study of the Mill Island ice core

• Main Author: Inoue, M
• Format: Thesis
• Language: English
• Published: 2016
• Subjects: Ice core; Mill Island; East Antarctica; Temperature reconstruction; Sea ice proxy; Climate proxy; Antarctic; Law Dome; Antarc*; Antarctica; ice core; Sea ice
• Online Access: https://eprints.utas.edu.au/23071/; https://eprints.utas.edu.au/23071/1/Inoue_whole_thesis.pdf

The IPCC 5th Assessment Report states that there are insufficient Southern Hemisphere climate records to adequately assess climate change in much of this region. Ice cores provide excellent archives of past climate, as they contain a rich record of past environmental tracers archived in trapped air and precipitation. However Antarctic ice cores, especially those from East Antarctica, are limited in quantity and spatial coverage. To help address this, a 120 m ice core was drilled on Mill Island, East Antarctica (65° 30' S, 100° 40' E). Mill Island is one of the most northerly ice coring sites in East Antarctica, and is located in a region with sparse ice core data. The specific project aims were: 1) To produce a high resolution, welldated record of water stable isotopes (δ18O, δD), and trace ion chemistry (sea salts, sulphate, methanesulphonic acid); 2) to investigate the seasonal and interannual variability of sea salts, in order to reveal which climate factors influence the Mill Island record; 3) to perform a regional comparison of δ18O and snow accumulation rate with nearby existing climate records from ice cores, observational stations, and atmospheric models, in order to seek the optimal method for temperature reconstruction using the Mill Island ice core record. Hydrogen peroxide, water stable isotopes, and trace ion chemistry were measured at high resolution throughout the entire core. The ice core was dated using a combination of chemical species, but primarily using water stable isotopes. The Mill Island ice core contains 97 years of climate record (1913 { 2009), and has a mean snow accumulation of 1.35 metres (ice-equivalent) per year (mIE/yr). Concentrations of trace ions were generally higher than at other Antarctic ice core sites (e.g., mean sodium levels were 254 μEq/L). The full trace ion record contained a mix of periods with well-defined seasonal cycles and periods with weak seasonality and a higher baseline. An abrupt change was observed in the sea salt record in the mid-1930s. This may be related to a significant change in the local ice-scape. Sea salts were compared with instrumental data, including atmospheric models and satellite-derived sea ice concentration, to investigate influences on the Mill Island ice core record. The mean annual sea salt record did not show a correlation with wind speed. Instead, sea ice concentration to the east of Mill Island was found to be likely influencing the annual mean sea salt record, at least during the period of 1979 to 2009. A mechanism involving formation of frost flowers on sea ice was proposed to explain the extremely high sea salt concentration. Post-depositional migration of magnesium and methanesulphonic acid were observed in the trace ion record, and for the first time, migration of sodium and chloride were observed. Snow accumulation rate was compared with snow accumulation or precipitation record from nearby sites. The Mill Island snow accumulation was found to be influenced by local orography, i.e., the annual snow accumulation record is not strongly related with precipitation in nearby sites. The Zonal Wave Three (ZW3), large scale atmospheric mode, modulates precipitation at nearby Law Dome, and to a lesser extent, modulates Mill Island precipitation. Snow accumulation and δ18O were compared with precipitation and temperature data from atmospheric models. The climatology of precipitation at Mill Island shows evidence of higher snowfall during winter, consistent with other Antarctic sites. The linear monthly ice core dating was adjusted using the precipitation climatology, and the adjusted δ18O record resulted in a warmer annual signal. This finding indicates that without this adjustment, there is a small cold bias in annual temperature reconstructions from ice cores that share this elevated winter precipitation. This bias should be considered when reconstructing temperatures where climate trends differ with season and when comparing with other temperature reconstructions (e.g., terrestrial or ocean based records). In situ temperature data (e.g., a co-located Automatic Weather Station) are not available at Mill Island. Instead, the annual mean δ18O record was compared with atmospheric reanalysis model output temperature at Mill Island. The correlation was found to be statistically insignificant. To attempt a more accurate palaeothermometer reconstruction, the annual record was divided into summer and winter \windows" with the maximum δ18O value set as the summer window centre, and the minimum δ18O value set as the winter window centre. It was found that when using narrow summer and winter windows, the δ18O value was significantly correlated with December to April mean model temperatures and May to July mean model temperatures, respectively.