Impact and Implications of Melting on the Preservation of Trace Elements in High-Alpine Snow and Glacier Ice

• Main Author: Avak, Sven Erik
• Format: Doctoral or Postdoctoral Thesis
• Language: English
• Published: Universität Bern
• Subjects: 540 Chemistry; Greenland; Antarc*; Antarctica; glacier; ice core
• Online Access: http://boristheses.unibe.ch/1356/; http://boristheses.unibe.ch/1356/1/19avak_s.pdf; https://doi.org/10.24442/boristheses.1356

Cold high-Alpine glaciers are invaluable archives of past climate and atmospheric composition. Especially trace element records from high-Alpine ice cores and snow pits contain comprehensive information about paleo atmospheric changes. Monitoring past environmental pollution is particularly important for Europe, one of the world’s most densely populated and most highly industrialized regions. Trace element records from different high-Alpine sites revealed that human activities have significantly impacted the composition of the atmosphere during the last 150 years. Unanswered questions still remain. For instance, the onset of European anthropogenic impact on the atmosphere, such as the impact from the earliest metal production in Western Europe interfering with natural background levels of trace elements from mineral dust deposition, has not been identified yet. Due to the current global climate warming, particularly pronounced for mountain regions such as the European Alps, many glaciers even at high altitudes are increasingly in danger to significantly suffer from melting. Apart from severe socioeconomic impacts caused by glacial melting, as Alpine glaciers are the major fresh water resource in Europe, meltwater percolation has been shown to substantially alter the information stored in these environmental archives. To further use trace element records as paleo atmospheric archives to investigate the open research questions, the influence of melting on the preservation of trace elements in snow and ice needs to be thoroughly understood. Only little and ambiguous information is available on meltwater-induced relocation of trace elements so far. The behavior of atmospheric impurities during meltwater percolation is assumed to be strongly dependent on their location in the ice microstructure. This spatial distribution of impurities at a grain scale is likely to be determined by rearrangement processes during snow metamorphism. However, information on the micro scale distribution of trace elements in Alpine snow and glacier ice and on the corresponding role of snow metamorphism is not yet available. In this thesis, part of the interdisciplinary “Microscale Distribution of Impurities in Snow and Glacier Ice (MiSo)” project, the behavior of trace elements during melting of high-Alpine snow and glacier ice was extensively investigated to assess their potential as reconstruction proxies in melt-affected ice core and snow pit records. Particular attention was dedicated to understand the underlying causes and mechanisms leading to the observed trace element behavior during melting, including the spatial distribution of trace elements in high-Alpine glacier ice and rearrangement processes during snow metamorphism. To examine the impact of melting on the preservation of trace elements of natural and anthropogenic origin, a 50 m segment of an ice core from upper Grenzgletscher, Switzerland, was analyzed for 35 trace elements using discrete inductively coupled plasma mass spectrometry. This segment included a 16 m section affected by meltwater percolation in the firn part. A fractionation depending on water solubility and location at the grain scale was observed. Ba, Ca, Cd, Co, Mg, Mn, Na, Ni, Sr, and Zn revealed significant concentration depletion, while Ag, Al, Bi, Cu, Cs, Fe, Li, Mo, Pb, Rb, Sb, Th, Tl, U, V, W, Zr, and the rare-earth elements (Ce, Eu, La, Nd, Pr, Sc, Sm, Yb) were well preserved. Trace elements likely to originate from insoluble minerals were found to be mostly preserved, even though typically enriched on grain surfaces. Immobility with meltwater percolation is a result of their insolubility in water. Trace elements linked to water-soluble particles revealed a variable meltwater-mobility. While trace elements occurring in ultra-low concentrations tend to be preserved due to incorporation into the ice lattice, abundant trace elements are prone to meltwater-induced relocation due to exceeded solubility limits in ice and consequent segregation to grain surfaces. The size of the corresponding ions was found to have a negligible effect. For ice cores from high-Alpine sites partially affected by melting, records of Ag, Al, Bi, Cu, Cs, Fe, Li, Mo, Pb, Rb, Sb, Th, Tl, U, V, W, Zr, and the rare-earth elements are proposed to be still applicable as robust environmental proxies. In collaboration with the Swiss Snow and Avalanche Research Institute, the impact of melting on the preservation of trace elements in snow was studied by conducting an extensive snow pit campaign at the Weissfluhjoch test site, Switzerland, with regular sampling from January to June 2017, to monitor the behavior of trace elements during melting of the snow pack. Comparison of snow pit profiles representing dry (insignificant occurrence of melting) and wet conditions (snow pack heavily soaked with meltwater) revealed a preferential loss of certain trace elements depending on their presumed microscopic location and their water solubility. The obtained elution behavior matched the findings from the upper Grenzgletscher ice core. Variable mobility was observed for trace elements originating from water-soluble particles, where low abundant trace elements were preferably retained. Concentration-independent preservation was visible for water-insoluble trace elements, owing to their meltwater immobility. Precipitation at the two 180 km distant high-Alpine sites upper Grenzgletscher andWeissfluhjoch is characteristic for Central European atmospheric aerosol composition. As the large majority of investigated trace elements revealed a consistent behavior with meltwater percolation at those two sites, the proposed applicability of trace elements as reconstruction proxies in melt-affected ice core and snow pit records is therefore most likely representative for the entire Alpine region. The redistribution of six major ions (ammonium, calcium, chloride, fluoride, sodium, sulfate) and 35 trace elements during artificial and natural snow metamorphism was extensively investigated in another collaboration with the Swiss Snow and Avalanche Research Institute. For this, artificial and natural snow samples were exposed to a controlled temperature gradient of 40 K m−1 in the laboratory for up to 90 days. Simultaneously, the distribution of the same atmospheric impurities was studied in samples taken from different depths of the snow pack at the Weissfluhjoch test site, each corresponding to a distinct exposure time of a natural temperature gradient. Initial snow structures, monitored by X-ray micro-tomography, and impurity distribution, determined by elution experiments, varied strongly between the different snow samples. However, with progressing snow metamorphism, snow structures became similar and ions exhibiting a high solubility in ice (ammonium, fluoride, chloride) were gradually buried in the ice interiors, whereas calcium, sodium, and sulfate were enriched at ice crystal surfaces. The redistribution of atmospheric impurities during snow metamorphism was shown to be strongly dependent on the temperature gradient, the exposure time, and the chemical composition. The observed preferred incorporation of certain species into the ice interior during snow metamorphism is correlated with their persistence during meltwater percolation. The elution experiments allowed investigation of water-soluble major ions only, whereas results for the trace elements could not be interpreted due to non-quantitative dissolubility of trace elements in the deployed eluent (ultra-pure water). An analytical method for the direct in situ analysis of trace elements at a submillimeter resolution in high-Alpine glacier ice was developed in collaboration with the Institute of Geochemistry and Petrology at ETH Zurich. This method is based on laser ablation inductively coupled plasma mass spectrometry. The development process comprised the construction and the consistent further development of a cooled sample holder, featuring an automatic coolant leakage detection system and compatibility to a commercially available laser ablation system, as well as choice of the optimal cooling medium, customization of the pre-existing laser ablation hardware and software, and the development of additional equipment for both sample preparation and handling. In addition to this, a measurement procedure for high-Alpine glacier ice was established, involving the determination of appropriate laser ablation parameters and setting up a procedure for signal intensity quantification. The availability of an internal standard in ice was evaluated and an approach to prepare matrix-matched ice standards from multi-element standard solutions for external calibration was established. The acidity of the multi-element solutions and the storage time of the ice standard after preparation were found to have the most significant impact on the calibration. Preliminary measurements of high-Alpine glacier ice samples from upper Grenzgletscher demonstrated that samples exhibiting an overwhelming mineral dust abundance do not provide evidence for a linkage between micro-scale distribution of trace elements and the grain boundary network. Such a dispersion of atmospheric contaminants in the ice matrix has also very recently been reported for layers with high impurity enrichment in deep ice from Antarctica and Greenland. Future work should involve in situ analysis of high-Alpine glacier ice exhibiting ultra-low levels of trace elements to minimize the influence of dust particles on the fractionation of trace elements at a grain scale and to further directly corroborate the indirect assessment of trace element location in the firn part of the ice core from upper Grenzgletscher. This requires further background suppression of the developed micro analytical method. The proposed applicability of trace elements as reconstruction proxies in melt-affected high-Alpine ice core and snow pit records should be reviewed for other regions with a different overall trace element composition, as high-mountain glaciers worldwide are increasingly affected by melting. For instance, the presence of water-insoluble trace elements, less prone to meltwater-induced relocation, is favored in glacier ice where higher mineral dust content prevails. Additionally, the impact of melting on the preservation of other reconstruction proxies, such as mercury or black carbon, should be investigated to possibly expand the set of rather “meltwater-persistent” proxies.