Testing the limits of Ti-in-Quartz thermometry and diffusion modelling to determine the thermal history of the Fish Canyon Tuff

How silicic magmas are stored in the upper crust before they erupt to form 100-1000s km3 ash-sheets remains a fundamental, but unanswered question in volcanology. While some studies posit storage in an eruptible state at low viscosity (<50% crystals) and high temperatures (>760-740°C) (warm storage), others suggest storage in a rigid state (>50% crystals) at lower, near-solidus temperatures (cold storage). Storage temperature and time spent near the solidus are typically constrained by mineral thermometry and diffusional relaxation modelling (at a given temperature), respectively. Since quartz is abundant over a range of temperatures and compositions and can incorporate titanium (Ti) at magmatic temperatures, a Ti-in-Quartz thermometer has been calibrated and Ti diffusion coefficients (DTi) have been measured. However, simply applying this thermometer or diffusion coefficient to volcanic quartz is burdened by an ongoing debate regarding their experimental calibration. This debate centers around three recent Ti-in-Quartz thermometers by Huang & Audétat (2012), Zhang et al. (2020), Osborne et al. (2022) and three DTi by Cherniak et al. (2007), Jollands et al. (2020), Audétat et al. (2021), each of which when applied to igneous systems favors either warm or cold storage. To determine their applicability for estimating the pre-eruptive thermal history of silicic magmatic systems, we apply the different Ti-in-Quartz thermometers and DTi to quartz from the Fish Canyon Tuff (USA). This tuff is an optimal location for such a study because it is a prime example of cold storage with multiple previous studies providing constraints on its storage conditions. We find that a temperature of 737 ± 16°C using the Zhang et al. (2020) thermometer is the most consistent with other temperature estimates for the Fish Canyon Tuff. Temperatures calculated using Huang & Audétat (2012) are acceptable, while those using Osborne et al. (2022) are unrealistic. Applying each of the DTi to quartz in the Fish Canyon Tuff and comparing these timescales to timescales from Ba-in-Sanidine diffusion and the total storage time of the mush (derived from the range in zircon U-Pb ages and the local eruption history), three different scenarios for pre-eruptive storage are possible. At a temperature of 737°C, timescales using DTi by Audétat et al. (2021) exceed the total storage time of the Fish Canyon system by ~2 Myr. These DTi are only consistent if storage temperatures were significantly higher, implying warm storage. Such a scenario is inconsistent with cold storage of the Fish Canyon system. Timescales derived from DTi by Jollands et al. (2020) and Cherniak et al. (2007) are consistent with cold storage of the Fish Canyon system. While DTi by Jollands et al. (2020) suggest long-term storage near 737°C and an extended period of pre-eruptive reheating, DTi by Cherniak et al. (2007) suggests storage below 737°C and rapid reheating.