MAGMA: Magma Accommodation and Ground Movement Analysis

岩浆将改变我们如何在火山地面运动的数值模型中捕捉火山下复杂的地质情况，这些模型用于火山爆发威胁评估；这将有助于在预测火山爆发方面取得重大进展，帮助保障人们的安全。超过8亿人居住在火山附近。为了保证人们的安全，我们需要可靠地预测火山爆发的时间，以及它们的潜在规模、形式和危害。然而，由于每个火山下的地质和管道系统动力学都是独一无二的，所有火山的行为都不同，这使得可靠地预测喷发变得困难。当岩浆侵入地壳并积聚时，它经常会推高上覆的岩石和地球表面。因此，活火山的地面运动往往预示着喷发。因此，我们利用卫星和地面工具监测火山的地表高度，寻找与岩浆积聚有关的地表运动迹象。利用复杂的数值模型，我们可以估计驱动测量到的地面运动所需的岩浆的数量和位置。这些对岩浆体的估计为喷发预测提供了重要的输入，因为它们限制了岩浆离地表的距离、岩浆的体积和压力以及岩浆移动的速度。许多地面运动模型的一个关键缺陷是，它们假设岩浆运动的岩石是简单的，没有内部结构或成分变化。然而，我们知道岩石在各种尺度上的物理和化学变化，以及它们是如何在空间和时间上变形的。关键的是，包含更真实地质（如分层）的地面运动模型表明，即使是很小程度的复杂性也会使估计的岩浆体性质发生数量级的变化。岩浆体估计的这种变化可能是预测喷发与否的区别。可靠地利用地面运动来预测火山爆发的时间、大小、风格和危险需要模型真实地捕捉宿主岩石的复杂性。例如，在注入岩浆之上的隆起需要上覆岩石弯曲。然而，我们实际上对岩石的抗弯曲能力知之甚少。我们也不知道弯曲岩石体积内的局部拉伸和压缩如何改变其材料特性，从而影响其对进一步变形的响应。至关重要的是，这些对岩石弯曲的控制决定了注入岩浆上方地面运动的程度和位置。为了在MAGMA中解决这些问题，我们将：1)进行力学实验，我们对不同的岩石进行加载和弯曲，以测量它们的弯曲阻力；实验室结果将使用经过试验和测试的岩土技术方法进行“放大”，以便它们能够代表整个岩体。2)研究地表暴露或三维地震反射数据成像的岩浆体上方古隆起区域的几何形状和变形，提供类似超声波的地下图像。3)野外采集样品，在实验室进行力学实验，对其进行拉压，测试局部弯曲如何改变其抗变形能力。MAGMA将利用这些信息建立新的有限元数值模型，该模型可以捕获宿主岩石复杂性的多个方面。这些模型将采用摄动理论方法，这种方法已经在复杂的地震地面运动模型中得到了成功的尝试和检验。利用我们开发的方法，我们将创建火山系统的合成但现实的模型，以模拟预先定义的岩浆体上方的地面运动。通过整合宿主复杂性和不同的输入，我们将测试不同的地质场景。我们的工作将使地面运动模型逐步发生变化，这将提高喷发预测的可靠性和对火山过程的理解，有助于保障人们的安全。

MAGMA will transform how we capture the complex geology beneath volcanoes within the numerical models of volcano ground movement that are used in eruption threat assessment; this will help drive significant improvements in forecasting eruptions, helping keep people safe and secure.Over 800 million people live near volcanoes. To keep people safe and secure, we need to reliably forecast when volcanic eruptions may occur and what their potential size, style, and hazards will be. However, because the geology and plumbing system dynamics beneath each volcano is unique, all volcanoes behaves differently, making it difficult to reliably forecast eruptions. As magma intrudes through the crust and accumulates, it often pushes up the overlying rock and Earth's surface. Ground movements at active volcanoes thus often herald eruption. We therefore monitor the surface elevation of volcanoes, using satellites and ground-based tools, to look for tell-tale ground movements related to magma build-up. Using sophisticated numerical models, we can estimate the amount and location of magma required to drive measured ground movement. These estimates of magma bodies provide crucial inputs for eruption forecasts as they constrain how close magma may be to the surface, its volume and pressure, and how fast it is moving. A key flaw of many ground movement models is that they assume the rocks through which magma moves are simple and have no internal structure or compositional variation. Yet we know rocks vary physically and chemically at all scales, and how they deform changes in space and time. Critically, ground movement models that include more realistic geology (e.g. layering) show that, incorporating even small degrees of complexity can change estimated magma body properties by orders of magnitude. Such changes in magma body estimates may be the difference between forecasting an eruption or not. Reliably using ground movement to forecast volcano eruption onset, size, style, and hazards requires models that realistically capture host rock complexity. For example, uplift above injecting magma requires the overlying rock to bend. Yet we actually know very little about how resistant rocks are to bending. We also do not know how local extension and compression within the bending rock volume changes its material properties and thus affects its response to further deformation. Critically, these controls on rock bending dictate how much and where ground movement occurs above injecting magma. To solve these problems in MAGMA, we will:1) Conduct mechanical experiments where we load and bend different rocks to measure their resistance to bending; lab results will be 'upscaled' using tried and tested geotechnical methods so they are representative of entire rock masses.2) Examine the geometry of and deformation within ancient areas of uplift above magma bodies exposed at the surface or imaged in 3D seismic reflection data, which provides ultrasound-like images of the subsurface.3) Collect samples from field areas and use mechanical experiments to pull and compress them in the lab to test how bending locally changed their resistance to deformation. MAGMA will use this information to build novel Finite Element numerical model that can capture multiple aspects of host rock complexity. These models will adopt a perturbation theory approach, which has been successfully tried and tested in complex ground movement models of earthquakes. With our developed method, we will create synthetic but realistic models of volcanic systems to simulate ground movement above pre-defined magma bodies. By incorporating host complexities and varying inputs, we will test different geological scenarios. Our work will enable a step-change in ground movement modelling, which will lead to improvements in the reliability of eruption forecasting and understanding of volcanic processes, helping keep people safe and secure.