What does the impurity variability at the microscale represent in ice cores? Insights from a conceptual approach

Abstract. Measuring aerosol-related impurities in ice cores gives insight into Earth's past climate conditions. In order to resolve highly thinned layers and to investigate post-depositional processes, such measurements require high-resolution analysis, especially in deep ice. Micron-resolution impurity data can be collected using laser ablation inductively coupled plasma mass spectrometry (LA-ICP-MS) but this requires careful assessment to avoid misinterpretation. 2D imaging with LA-ICP-MS has provided significant new insight, often showing an association between soluble impurities and the ice crystal matrix, but interpreting 1D signals collected with LA-ICP-MS remains challenging partially due to this impurity-boundary association appearing in the high-frequency component of signals. In this work, a computational framework has been developed integrating insights from 2D imaging to aid the interpretation of 1D signals. The framework utilises a simulated model of a macroscopic ice volume with a representative microstructure and soluble impurity localisation that statistically represents distributions seen in 2D maps, allowing quantitative assessment of the imprint of the ice matrix on 1D signals collected from the volume. Input data were collected from four ice core samples from Greenland and Antarctica. For the samples measured, quantifying the variability of 1D signals due to the impurity-matrix imprint shows that modelled continuous bulk signal intensity at the centimetre scale varies below 2 % away from an idealised measurement that captures all variability. In contrast, modelled single-profile micron-resolution LA-ICP-MS signals can vary by more than an average of 100 %. Combining individual LA-ICP-MS signals into smoothed and spatially averaged signals can reduce this variation to between 1.5 and 5.9 %. This approach guides collecting layer-representative signals from LA-ICP-MS line profiles and may help to bridge the scale gap between LA-ICP-MS data and data collected from meltwater analysis.