Falling Basins: revealing hidden faults from patterns of land subsidence from water extraction using Earth Observation data

世界各地仍有许多地震断层有待发现，有时是因为它们隐藏在地形中，被盆地中的沉积物所覆盖。城市建在这些盆地上，因为它们为农业提供了肥沃的土壤，并提供了地下的水源。然而，抽取地下水会引起严重的地面沉降。虽然这是住房和供水的问题，但它提供了一个重要的机会来发现城市地下的断层，这些断层可能会造成未来的地震风险。我们可以探测到这些断层，因为它们优先控制地下水在地下和含水层内的流动，因此对地表下沉的方式表现出结构性控制。该项目将使用最新的地球观测卫星，如Sentinel-1，利用InSAR技术检测这些沉降模式，以找到世界各地主要城市下方的隐藏断层（Elliott, 2020）。我们预计，包括水平和垂直运动将改善沉降和断层位置的检测。这项工作很重要，因为地震危险和隐藏的断层可能会影响到发展中的城市中的许多人（Crowley & Elliott, 2012）。许多主要断层位于山脉和盆地的边缘，并容纳了这两个构造域之间的相对运动。然而，随着变形的时间迁移，盆地内部也会形成断层，断层可能隐藏在盆地内部（Elliott et al., 2020）。这项研究将是跨学科的，并直接与灾害风险和人道主义从业人员接触，以定制相关的基于科学的产出，以便将其纳入灾害风险评估和应急情景规划。因此，这将有助于实现联合国关于可持续城市的发展目标，并减少灾害造成的生命损失。通过孔隙弹性建模（Gambolati & Teatini, 2015），可以更好地了解地下水抽取的地质力学影响和流体流动的断层控制，从而限制人为地面沉降的模式。现在，利用诸如Sentinel-1之类的卫星雷达，可以在观测区域上空的时空形变模式方面实现阶梯式的变化，其精度比利用InSAR每年提高几毫米。该项目将利用这些基于空间的不同沉降率测量来测试揭示活跃变形地区城市地下隐藏断层的方法。它将测试这样一个假设，即通过直接图像和计算机视觉分析观测结果以及含水层流动的数值模型，除了使用垂直速率外，还使用水平变形大大增强了断层检测。通过量化沉降的时空格局，利用独立成分分析等数据分析技术，识别沉积物中潜在的隐藏断层将成为可能。通过将其与盆地范围内的沉降压实模型预测的地表变形进行比较，可以检测到作为流体流动屏障或管道的断层，因为这些断层会改变一阶沉降信号。一旦确定，相对于暴露的城市人口，断层的位置和大小将与主要利益相关者共同设计，使用基于场景的方法进行应用地震危害和风险分析（Hussain等人，2020）。该学生将在利兹大学地球与环境学院地球物理与构造研究所活动构造组的John Elliott博士的指导下工作。该项目将由Mark Thomas博士（也在IGT， SEE）和Kate Crowley博士（爱丁堡大学）共同监督。利兹的地球物理和构造研究所还主持了地震、火山和构造观测和建模中心（COMET），该中心提供了大量从事活动技术的研究人员

Many earthquake faults remain to be discovered around the world, sometimes because they are hidden in the landscape, covered by sediments in basins. Cities are built on these basins because they offer fertile ground for agriculture and sources of water contained within the ground. However, groundwater extraction can induce significant land subsidence. Whilst a problem for housing and water supply, it provides an important opportunity to find faults beneath the city that may pose a future earthquake risk.We can detect these faults as they preferentially control how ground water flows in the subsurface and within aquifers, and consequently exhibit a structural control on how the ground surface sinks. Using the latest Earth Observation satellites, such as Sentinel-1, this project will detect these subsidence patterns to find the hidden faults beneath major cities around the world using the technique of InSAR (Elliott, 2020). We anticipate that including horizontal as well as vertical motion will improve detection of subsidence and fault locations. This work is important because seismic hazard and hidden faults potentially affect many people in growing cities (Crowley & Elliott, 2012). Many major faults lie along the edge of mountains and basins, and accommodate the relative motion between these two tectonic domains. However, faults also form within basins as the deformation migrates through time and the fault may be hidden within the basin (Elliott et al., 2020). This research will be interdisciplinary and engage directly with disaster risk and humanitarian practitioners to tailor relevant science-based outputs for integration within disaster risk assessment and emergency scenario planning. This will therefore support the realisation of UN development goal of sustainable cities and reduced loss of life from hazards.An improved understanding of the geomechanical effects of groundwater withdrawal and the control of faults of the flow of fluids will be gained through poro-elastic modelling (Gambolati & Teatini, 2015) placing constraints on the patterns of anthropogenic land subsidence. A step change in observing spatial and temporal patterns of deformation over regional areas is now possible with satellite radar such as Sentinel-1, with accuracies better than a few millimetres per year using InSAR. This project will use these space-based measurements of differential subsidence rates to test methods that reveal these concealed faults beneath cities in actively deforming regions. It will test the hypothesis that fault detection is greatly enhanced with the use of horizontal deformation in addition to vertical rates, both through direct imagery and computer vision analysis of the observations and in numerical models of aquifer flow. By quantifying the spatial and temporal pattern of subsidence, using such data analysis techniques as Independent Component Analysis, identification of potential hidden faults within the sediments will be possible. By comparing this to predictions of surface deformation from basin-wide compaction modelling of subsidence, the faults acting as barriers or conduits to fluid flow will be detected as these will alter the first-order subsidence signal. Once identified, the locations and sizes of faults relative to exposed urban populations will be co-designed with key stakeholders to produce applied seismic hazard and risk analyses using a scenario-based approach (Hussain et al., 2020).The student will work under the supervision of Dr. John Elliott, within the Active Tectonics group of the Institute of Geophysics & Tectonics in the School of Earth & Environment at Leeds. The project will be co-supervised by Dr Mark Thomas (also in IGT, SEE) and Dr Kate Crowley (University of Edinburgh). The Institute of Geophysics & Tectonics at Leeds also hosts the Centre for the Observation and Modelling of Earthquakes, Volcanoes and Tectonics (COMET) which provides a large group of researchers engaged in active tec