DataSheet4_Pre-eruptive excess volatiles and their relationship to effusive and explosive eruption styles in semi-plugged volcanoes.pdf

Understanding the role of various factors influencing eruption style is challenging, but it can aid in adapting different hazard mitigations and crisis responses for explosive or effusive events. Here, we focus on the role of magma storage conditions in controlling eruption styles at basaltic andesite volcanoes, and how they can be related to monitoring data. We study the cycle of explosive (1990, sub-Plinian) → effusive (2007, dome) → explosive (2014, sub-Plinian) eruptions from Kelud (Kelut) volcano, Indonesia. We conducted petrological analyses of the eruption products and phase equilibria experiments using pumice and explored a range of temperatures, pressures, oxygen fugacity, and volatile contents. We show that we can reproduce the main mineral assemblage (plagioclase ± pyroxenes ± magnetite ± amphibole ± olivine) and phenocryst content (30–50 wt%) of the magmas from the three eruptions at T = 975 ± 39°C, p = 175 ± 25 MPa, fO2 = nickel–nickel oxide buffer, and about 4–6 wt% water in the melt (ca. 3 to 5 wt% and ca. 4 to 7 wt% for the 1990 and 2014 eruptions, respectively). However, geothermobarometric results also indicate that some crystals of amphibole were sourced from higher pressures. We infer from a synthesis of our data and historical observations that the high phenocryst content of the 2007 dome (∼70 wt%) likely resulted from slow magma ascent toward the surface alongside progressive degassing and re-equilibration at a lower volatile content (∼1 wt% water in the melt). Mass balance calculations on the sulfur budget of the 1990, 2007, and 2014 magmas show that the explosive events contained an excess fluid phase at pre-eruptive conditions, and we propose that this led to their higher explosivity compared to the 2007 dome. The accumulation of excess fluids during decadal-long repose depends on how plugged the volcanic system is, or its ability to passively release magmatic fluids, prior to eruption. Such condition could be inferred from monitoring records, including changes in gravity of the plumbing system over time, and thus contribute to better anticipate eruptive style.

厘清影响火山喷发样式的各类因素的作用机制颇具挑战，但相关认知可为爆裂式与溢流式火山事件制定差异化灾害减缓策略及危机响应方案提供支撑。本研究聚焦玄武安山岩（basaltic andesite）火山的岩浆存储条件对喷发样式的控制作用，及其与监测数据的关联机制。我们以印度尼西亚克卢德（Kelut）火山的喷发旋回为研究对象：1990年亚普林尼式（sub-Plinian）爆裂喷发→2007年穹丘状溢流式喷发→2014年亚普林尼式爆裂喷发。我们以浮岩为研究对象，开展了喷发产物的岩石学分析（petrological analyses）与相平衡实验（phase equilibria experiments），探究了温度、压力、氧逸度及挥发分含量的变化范围。研究结果表明，在温度975±39℃、压力175±25MPa、氧逸度为镍-镍氧化物（nickel–nickel oxide）缓冲对条件下，熔体水含量约4~6wt%（1990年喷发熔体水含量约3~5wt%，2014年喷发约4~7wt%）时，可复现三次喷发岩浆的主要矿物组合（斜长石（plagioclase）±辉石（pyroxenes）±磁铁矿（magnetite）±角闪石（amphibole）±橄榄石（olivine））与斑晶含量（phenocryst content，30~50wt%）。不过，温压地球化学（geothermobarometric）分析结果显示，部分角闪石晶体源自更高压力的岩浆环境。结合本研究数据与历史观测结果综合推断，2007年穹丘喷发的高斑晶含量（约70wt%）可能源于岩浆缓慢上升至地表的过程中，伴随渐进式脱气并在较低挥发分含量（熔体水含量约1wt%）条件下发生平衡调整。对1990、2007及2014年岩浆的硫预算（sulfur budget）开展质量平衡计算（mass balance calculations）后发现，爆裂式喷发在喷发前条件下存在过剩流体相，据此我们认为这是其相较2007年穹丘喷发具有更高爆发性的原因。十年尺度休眠期内过剩流体的积累，取决于火山系统的堵塞程度，或是喷发前被动释放岩浆流体的能力。这类条件可通过监测记录（包括岩浆通道系统（plumbing system）随时间的重力变化）进行推断，从而助力更精准地预判喷发样式。