This thesis investigates two problems in the field of volcanology by using a combination of remote sensing of volcanic gases, magma modelling and fluid dynamics. The first problem relates to gas-flow through particle-liquid suspension; with relevance for understanding the mechanism of intermittency and chemical changes in the gas composition in basaltic eruptions. The second problem centres around turbulent structures in vertical and wind-blown plumes. First, the injection of gas into a cylinder, which was filled with a water-glycerol mixture and particles, was investigated. These analogue experiments provides the framework of a model for episodic slug formations. The gas pressure at the base of the conduit is large enough to overcome the yield stress of crystal-rich magma and form gas channels within the pack. At a critical depth, the gas may displace the overlying crystal-rich plug en masse, producing a growing slug of gas. This plug becomes progressively sheared and weakened until gas enriched in the least soluble volatiles break through, causing an explosion at the surface enriched in CO2 and SO2. When the gas slug ascends, melt is drawn up in its wake, which exsolves the more soluble volatile components like HCl during passive degassing until the next explosive slug-bursting event. The second problem explores the turbulent structures of volcanic plumes. I show that it is possible to estimate volcanic water vapour fluxes with the combination of the measurement of convective structures in plumes and classical plume theory. This model predicts an estimate of H2O flux of Villarrica within 18 and 30 kg s-1 and between 130 and 270 kg s-1 for Etna, assuming the volcanic gas temperature is in the range of 500 to 900 ̊C. These estimated gas fluxes are reasonable estimations of fluxes derived from spectroscopy combined with electrochemical sensors of previous studies.