Looking inside the Continents from Space: Insights into Earthquake Hazard and Crustal Deformation

当两个构造板块一起或分开移动时，被困在它们之间的任何大陆都会变形，导致主要地质特征，如山脉带或沉积盆地的发育。当脆性的近地表地壳试图容纳这种变形时，地震发生在地球内部的断层上。迫切需要了解大陆如何变形，以及地震将在哪里发生--自1900年以来，已有140万至170万人死于大陆内部的地震。我们可以使用卫星测量大陆正在积极变形的方式。GPS可以非常精确地测量地面上的各个点是如何移动的，但这些点通常是稀疏分布的。在过去的二十年里，欧洲航天局(欧空局)设计的卫星展示了星载雷达测量地球表面位移的能力。雷达反复发出一连串的微波信号，从表面散射回来，并在返回航天器时进行测量。我们使用卫星在两个不同时间获取的雷达回波的差异来测量该点在中间时间间隔内的位移。几毫米或更小的位移可以用这种方法测量。随着大陆地壳的变形，岩石继续弯曲，形成应变，将在未来的地震中释放出来。在评估地震危险性时，除了知道将发生地震的断层在哪里外，还必须知道这种应变的增长速度。然而，这些速率很小，在存在由雷达散射的地面和雷达信号通过的大气特性变化引起的噪声的情况下，使用雷达不容易测量。此外，我们对卫星位置的了解中的错误会影响我们的测量。可以设计出应对这些困难的方法，但由于在地震带的许多地区不定期和不频繁地获取雷达图像，目前的卫星应用这些方法的机会有限。由于欧空局计划在2013年中末发射一颗新的雷达卫星Sentinel-1A，我们有动力让一个调查小组努力解决这些问题。一个相同的合作伙伴，哨兵-1B将在18个月后发射。每个航天器将每6天经过地球表面的一个给定点；一旦两个航天器都进入轨道，任何一个点都将每3天重新访问一次。这一短暂的时间间隔，加上航天器的每一次飞行都将进行观测，其位置将受到仔细控制和众所周知，这将意味着我们测量运动速度和应变的能力将得到根本改善。通过将所有可用的卫星轨道的测量结果与任何可用的GPS数据相结合，我们将能够在大范围内详细绘制应变正在积累的速率地图。我们计划通过使用这样的观测来测试和约束物理模型，来观察大陆内部发生的变形。因此，用雷达测量的地震中发生的位移可以用来推断断层在深部发生的运动。地震附近的地球表面在告诉我们周围地区的力学性质后立即继续移动的方式，这些知识对于理解断层周围的力如何随时间变化至关重要。在更大的尺度上，大陆应变的空间分布告诉我们地壳强度的变化。有了这些限制，我们就可以检验关于大陆如何变形以及控制变形发生地点的主要因素是什么的相互矛盾的假设。

As two tectonic plates move together or apart, any continent trapped between them deforms, causing major geological features such as mountain belts or sedimentary basins to develop. As the brittle, near-surface crust tries to accommodate the deformation, earthquakes occur on faults inside the earth. The need to understand how the continents deform, and where earthquakes will occur, is compelling - between 1.4 and 1.7 million people have died in earthquakes in the continental interiors since 1900.We can measure the way the continents are actively deforming using satellites. GPS can provide very precise measurements of how individual points on the ground move, but such points are often sparsely distributed. Over the past two decades, satellites designed by the European Space Agency (ESA) have demonstrated the ability of satellite-borne radar to measure displacements of the earth's surface. The radar repeatedly sends out bursts of a microwave signal that scatters back from the surface and is measured when it returns to the spacecraft. We use differences in the radar returns acquired by the satellite at two different times to measure the displacement of that point over the intervening time interval. Displacements of a few millimeters or less can be measured in this way.As the continental crust deforms, the rocks continue to bend, building up strain that will be released in future earthquakes. When assessing earthquake hazard, in addition to knowing where the faults are on which the earthquakes will occur, it is essential to know the rate at which this strain is growing. These rates are small, however, and not easy to measure using radar in the presence of noise caused by changes on the ground from which the radar scatters and in the properties of the atmosphere through which the radar signal passes. In addition, errors in our knowledge of the position of the satellites affect our measurements. Methods can be devised to counter these difficulties, but the opportunities to apply them has been limited with the current satellites by the irregular and infrequent acquisition of radar images over many parts of the seismic belts.We are motivated to bring the efforts of a team of investigators to bear on these questions because of the planned launch by ESA in mid-to-late 2013 of Sentinel-1A, a new radar satellite. An identical partner, Sentinel-1B will be launched 18 months later. Each spacecraft will pass over a given point on the earth's surface every 6 days; once both are in orbit any point will be revisited every 3 days. This short time interval, plus the fact that observations will be made for every pass of the spacecraft and its position will be carefully controlled and well known, will mean a radical improvement in our ability to measure rates of motion and strain. By combining the measurements from all available satellite tracks, together with any GPS data available, we will be able to map in detail over large areas the rates at which strain is building up. We plan to look at what happens inside the continents as they deform by using such observations to test and constrain physical models. Thus the displacements occurring in an earthquake measured by radar can be used to infer the movements that have taken place on the fault at depth. The way the earth's surface in the vicinity of an earthquake continues to move immediately after it tells us about the mechanical properties of the surrounding region, knowledge essential to understanding how the forces around a fault vary with time. On a larger scale, the spatial distribution of strain in the continents tells us about changes in the strength of the crust. With these constraints we can test competing hypotheses about how the continents deform and what are the major factors controlling where the deformation occurs.