Improving magmatic CO2 reconstruction using X-ray Computed Tomography to accurately quantify melt inclusion volumes and geometries

Melt inclusions provide valuable insights into magmatic systems, allowing the study of otherwise inaccessible melts. Trapped volatile contents, commonly in the form of bubbles within melt inclusions, allow direct sampling and study of magmatic volatiles that are otherwise lost due to saturation and degassing. Quantifying magmatic volatiles such as CO2 is a complex process requiring multiple different analytical methods to extract volatile contents from poly-phase inclusions (glass and vapour bubble). These techniques use inclusion and bubble volumes to generate CO2 contents. At present, melt inclusion and bubble volumes are usually determined using 2D optical measurements, assuming a simple geometry and requiring an estimate for the third dimension. This study applies X-Ray Computed Tomography (XCT), a non-destructive, three-dimensional imaging technique, to produce more representative melt inclusion volume measurements for a large suite of olivine-hosted melt inclusions. We compare this method with conventional two-dimensional methods for assuming inclusion geometry and volume, showing that volume quantification can be significantly improved by using XCT. At best, optical methods are likely to overestimate bubble volumes by 14-40%, however, this is dependent on a multitude of factors that are likely to significantly increase these errors. Adopting a three-dimensional approach, XCT both improves the accuracy of inclusion volumes whilst also allowing for the determination of uncertainty, using repeat analysis and variable processing. This allows for better accuracy of inclusion volume estimates, ultimately improving CO2 reconstructions that are made from these measurements. In general, overestimation of melt inclusion volumes results in an underestimate of bubble CO2 concentrations generated from melt inclusion analysis. The improvement afforded by including XCT melt inclusion studies allows for better data and ultimately, more correct interpretation of melt inclusion-derived data, such as magmatic volatile contents, magma storage depths and pressure-temperature conditions.