Determining magma storage depths beneath volcanic systems is vital for interpreting geophysical signals of unrest at hazardous volcanoes, and to understand the structure and dynamics of volcanic systems. However, popular methods such as mineral thermobarometry have errors spanning the entire thickness of the crust in many oceanic settings (+-10-20 km). More precise methods such as melt inclusion barometry are time consuming and expensive, and as a result, have not be utilized as part of the syn-eruptive workflow of volcano observatories to understand evolving hazards. Here, we present Raman spectroscopy analysis of fluid inclusions (FIs) as a precise, fast and cheap method to determine magma storage. We show that minimal sample preparation requirements, rapid analysis and automated data processing mean that this method can even be used to determine storage depths during volcanic eruptions, to help intepret ongoing activity and make operational decisions.
FIs form when exsolved fluids within a volatile-saturated volcanic system get trapped within growing crystals. In systems with CO2-rich exsolved fluids (e.g., Hawai’i, Iceland, Samoa), these inclusions provide a very precise barometer, because the density of CO2 is strongly related to the pressure at which the fluids were entrapped through the CO2 equation of state. Recent advances in quantitative measurements of CO2 density by Raman spectroscopy and software tools for fitting spectral data and performing equation of state calculations mean that pressures can now determined accurately and quickly from fluid inclusions. Comparison of Raman measurements to densities obtained by microthermometry, and fluid inclusion pressures to melt inclusion saturation pressures, show that Raman measurements are within uncertainty of these more established approaches. Comparisons of fluid inclusion pressures to those obtained by mineral thermobarometry in Iceland and the Galápagos highlights the higher precision of fluid inclusions over these popular techniques. After validating the method, a rapid response simulation was performed during the Sept 2023 eruption of Kīlauea in collaboration with the Hawaiian Volcano Observatory, demonstarting that within a day of sample receipt, we could determine that magma was being supplied from the shallower reservoir at 1- 2 km depth. Through a compilation of the composition of exsolved fluids globally, it is apparent that the fluid inclusion method is highly suitable for application at many of the world’s most active volcanoes (Iceland, Hawaiʻi, Galápagos Islands, East African Rift, Réunion, Canary Islands, Azores, Cabo Verde), and could supplement other rapidly advancing petrological monitoring techniques. Finally, we present new constraints on magma storage beneath Hawaiian volcanoes at a wide variety of life stages (Kamaʻehuakanaloa, Kīlauea, Mauna Loa, Mauna Kea, Hualālai), tracking systematic changes in magma storage in response to a waxing and waning supply from the mantle plume.