Abstract
Large volcanic eruptions are frequently triggered by the intrusion of hot primitive magma into a more evolved magma-chamber or -mush zone. During intrusion into the cooler mush zone, the basaltic magma undergoes crystallization, which in turn releases heat and volatiles to the mush. This should cause a drop in bulk mush-viscosity, potentially leading to its mobilization and even eruption. The non-linear changes in the transport properties of both magmas during this interaction modulate how the magmas accommodate deformation during both interaction and ascent. As such, this interaction represents a complex disequilibrium phenomenon, during which the material properties guiding the processes (dominantly viscosity) are in constant evolution. Such a scenario highlights the importance of non-isothermal sub-liquidus processes for the understanding of natural magmatic and volcanic systems and underlines the need for a rheological database to inform on, and to model, this interaction process. Here we present new experimental data on the disequilibrium rheology of the least evolved end-member known to be involved in magma mixing and eruption triggering as well as lava flow processes in the Phlegrean volcanic district (PVD). We measure the melt´s subliquidus rheological evolution as a function of oxygen fugacity and cooling rate and map systematic shifts in its rheological “cut off temperature; Tcutoff” (i.e. the point where flow ceases). The data show that 1) the rheological evolution and solidification behavior both depend on the imposed cooling-rate, 2) decreasing oxygen fugacity decreases the temperature at which the crystallization onset occurs and modifies the kinetics of melt crystallization and 3) the crystallization kinetics produced under dynamic cooling are significantly different than those observed at or near equilibrium conditions. Based on the experimental data we derive empirical relationships between the environmental parameters and Tcutoff. These empirical descriptions of solidification and flow may be employed in numerical models aiming to model lava flow emplacement or to reconstruct the thermomechanical interaction between basalts and magma mush systems. We further use the experimental data in concert with models of particle suspension rheology to derive the disequilibrium crystallization kinetics of the melt and its transition from crystallization to glass formation.
Highlights
General OverviewThe growing number of inhabitants, tourists, and economic activities near volcanoes require adequate volcanic hazard assessment and mitigation plans to guide decision-making in the case of volcanic unrest
The results presented here show that cooling rate is the most important parameter in determining when and how solidification, and therewith heat and volatile transfer, occurs during cooling of basaltic melts
Since the Tcutoff model describes a crystallization induced threshold, it is important to highlight that it is only applicable at disequilibrium conditions and sub-liquidus temperatures. This Tcutoff model is bound by two distinct thermodynamic points: (1) the melts’ liquidus temperature at the high temperature end, above which no crystallization occurs and (2) the melts’ glass transition temperature (Tg) at the low temperature end, where solidification occurs by glass formation, not crystallization
Summary
The growing number of inhabitants, tourists, and economic activities near volcanoes require adequate volcanic hazard assessment and mitigation plans to guide decision-making in the case of volcanic unrest. Measurements at disequilibrium conditions, employing new experimental methods (Giordano et al, 2007; Kolzenburg et al, 2017, 2018a; Vetere et al, 2019) and infrastructure (Kolzenburg et al, 2016, 2018b) have only recently become more numerous This underlines the relevance of developing a comprehensive database of the temperature dependent rheology of crystallizing magmas and lavas in dynamic temperature–, shear-rate–, and oxygen fugacity (fO2) -space to accurately constrain physical property based lava flow and magma mixing models. We do this because sampling the interstitial melt composition is not possible in disequilibrium experiments due to the extremely rapid crystallization kinetics; see Kolzenburg et al (2017) and Kolzenburg et al (2018a) for details Following this approach, we find that the viscosity of the initial composition and the fractionated melt at equilibrium, when attaining 60 vol % crystallinity vary within less than ∼0.4 log units and this variation is smaller at higher temperatures. We can exclude that a change in melt composition has a drastic effect on the viscosity measurement and assign the relative viscosity changes over the experiment to crystal nucleation and growth (i.e., increasing suspension solid fraction)
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Free) 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
