Collaborative Research: The Provenance of the Excess Sulfur Released in Arc Volcanoes During Plinian Eruptions

合作研究：普林尼火山爆发期间弧状火山释放的过量硫的来源。虽然人们普遍认为岩浆中挥发性元素的行为控制着喷发动力学，但对浅岩浆系统中蒸汽的深度和析出量以及蒸汽泡的分布和输运仍然知之甚少。该项目旨在开发一种新的多学科方法来量化蒸汽脱气发生的位置和数量，蒸汽泡在不同岩浆环境中运输的速度，如果它们预计会积聚在系统的较浅部分，以及这种积聚将对喷发动力学产生多大影响。需要解释的一个基本观察结果是，在爆炸性喷发期间释放的挥发物与喷发前可以溶解在熔体中的挥发物的数量（过量脱气）之间存在差异。过量脱气最明显的例子与S质量平衡有关，因为喷发期间和之后的S释放可以通过光谱方法精确测量。几个悬而未决的问题将得到解决：(1)在喷发前岩浆中可以存在多少蒸汽，以及有多少蒸汽是在原位释放的，而有多少蒸汽是从更深的未喷发岩浆（非原位）中输送过来的？(2)如果离地脱气是重要的，那么是什么控制了岩浆体最浅层气泡的聚集，它又如何影响岩浆的喷发行为？这项工作具有变革性，因为它将利用pi开发的新数值模型探索岩浆系统中暴露挥发物的运输和储存，并将地球化学数据（磷灰石和磁黄铁矿中的微量元素）与热力学建模（预期蒸汽含量作为结晶度的函数）和流体动力学建模结合起来，形成一种量化岩浆中挥发物平衡的新方法。现有的热力学模型将包括H2O（如熔融体）和H2O- co2（如挥发性钙）的溶解度，在合理的初始条件范围内，确定具有记录的S过量的普林尼火山喷发（1980年的圣海伦火山，1982年的埃尔奇春火山，1991年的皮纳图博火山，1883年的喀拉喀托火山，1912年的卡特迈火山，公元181年的陶波火山）以及更大，更古老的富S单元（160公里的科斯高原凝灰岩）的溶解蒸汽百分比。鱼峡谷凝灰岩28 Ma)，其中大气负荷约束较差。在相同的单元中，通过电子探针对磷灰石和磁黄铁矿的已发表数据和/或分析将用于确定(a)喷发前岩浆在其浅层储层中是否蒸汽饱和（使用我们开发的地球化学测试）和(b)使用磷灰石的分带剖面，喷发前挥发性元素含量的变化情况。最后，将使用不同的数值技术开发多相流模型，以估计岩浆系统中移动蒸汽泡的情况以及喷发前它们在储层中积聚的速度。这些模型将为火山喷发动力学提供宝贵的新见解。更广泛的影响。该项目将为研究生提供支持，并将在一个高度多学科和固有定量的环境中培养他/她，使他/她具备现代科学的基本技能。它将提供一名新的研究员（Huber，没有先前的资助）资金，以发展一个活跃的物理火山学小组，并将允许两名年轻教员继续积极的多学科合作，这种合作在过去几年中取得了成功。最后，pi建议为高级本科生和研究生开发一门短期课程，将建模技术和地球化学分析应用于岩浆过程。在过去的两年中，PI Huber一直在加州大学伯克利分校、日内瓦大学（瑞士）和莱斯大学教授这门课程的一个版本，作为为期一周的系列讲座和实践练习。根据学生对这门课的热烈反应，我们期望在佐治亚理工学院和华盛顿大学举办这样的课程可能会吸引许多当地学生，并将有力地帮助我们的社区更加了解和精通地球科学的定量方法。

Collaborative Research: The provenance of the excess sulfur released in arc volcanoes during Plinian eruptionsIntellectual merit. Although it is commonly accepted that the behavior of volatile elements in magmas controls eruption dynamics, the depth and amount of vapor exsolved, as well as the distribution and transport of vapor bubbles in shallow magmatic systems remain poorly constrained. This project aims to develop a novel multidisciplinary methodology to quantify where and how much vapor degassing occurs, how fast vapor bubbles are transported in different magmatic environments, if they are expected to accumulate in the shallower portion of the system and how much effect this accumulation will have on eruption dynamics. A fundamental observation to be explained is the discrepancy between emitted volatiles released during explosive eruptions and the amount of volatiles that can be dissolved in the melt prior to eruption (excess degassing). The clearest example of excess degassing is associated with S mass balance, because S release during and after eruptions can be measured accurately by spectroscopic methods. Several open questions will be addressed: (1) How much vapor can be present in a magma prior to an eruption and what fraction was exsolved in-situ versus transported from deeper unerupted magma (ex-situ)? (2) If ex-situ degassing is important, what controls the accumulation of bubbles in the shallowest part of the magma body and how does it affect the eruptive behavior of the magma? The proposed work is transformative in that it will explore the poorly understood transport and storage of exsolved volatiles in magmatic systems with new numerical models developed by the PIs, and it will integrate geochemical data (trace elements in apatites and pyrrhotite) with thermodynamic modeling (expected vapor content as function of crystallinity) and fluid dynamics modeling into a novel approach to quantify the volatile budget in magmas. Existing thermodynamic models will be used that include H2O (e.g., MELTS) and H2O-CO2 (e.g., VolatileCalc) solubilities to determine, for a range of reasonable initial conditions, the percentage of exsolved vapor in given examples of Plinian eruptions with documented S excess (Mount St. Helens 1980, El Chichon 1982, Pinatubo 1991, Krakatau 1883, Katmai 1912, Taupo 181 AD) as well as for larger, older S-rich units (Kos Plateau Tuff 160 ky, and Fish Canyon Tuff 28 Ma) for which the S atmospheric load is poorly-constrained. In the same units, published data and/or analyses of apatites and pyrrhotites by electron microprobe will be used to determine (a) whether the magmas were vapor-saturated in their shallow reservoirs prior to eruption (using a geochemical test we have developed) and (b) using zoning profiles in apatites, how volatile element contents vary prior to eruption. Finally, multi-phase flow models will be developed using different numerical techniques to estimate how mobile vapor bubbles are across the magmatic system and how quickly they accumulate in reservoirs before an eruption. Such models will provide invaluable new insights in the dynamics of volcanic eruptions. Broader impacts. This project will support a graduate student, and will train her/him in a highly multi- disciplinary and inherently quantitative environment, enabling him/her with those fundamental skills of modern science. It will provide a new investigator (Huber, no prior funding) funds to develop an active group in physical volcanology and will also allow two young faculty members to continue an active multi-disciplinary collaboration that has been successful over the last few years. Finally, the PIs propose to develop a short course for advanced undergraduate and graduate students on merging modeling techniques and geochemical analyses applied to magmatic processes. PI Huber has been teaching a version on this course, over the last two years, as a week-long series of lectures and practical exercises at the University of California, Berkeley, the University of Geneva (Switzerland), and at Rice University. On the basis of the enthusiastic response of the students to this class, we expect that organizing such a course at Georgia Tech and University of Washington is likely to attract many local students, and would strongly help our community to become more aware and proficient in quantitative methods in Earth Sciences.