Understanding the physical processes causing surface creep on faults.

自1966年帕克菲尔德地震后观察到圣安德烈亚斯断层蠕动以来，人们就知道断层在地表滑动或蠕动。自那以后，圣安德烈亚斯断层一直沿着圣胡安包蒂斯塔和帕克菲尔德之间150公里长的区段爬行。自从在圣安德烈亚斯断层的这一段发现蠕动以来，已经安装了蠕变仪和对准阵列来监测蠕动2。在世界各地的许多其他断层系统上也观察到了地表蠕动。在其他大型走滑断层上也观察到了蠕动，如土耳其伊斯梅帕萨附近的北安纳托利亚断层3和莱特岛上的菲律宾断层4。走滑断层并不是唯一具有地表蠕动的断层类型，在台湾5，6反向断层上观察到蠕动。断层上的蠕动发生在称为蠕变事件的爆发中；这些可以在蠕变事件的记录中被识别(图1)。由于引起蠕动事件的物理过程，蠕动事件在滑动特征(滑动速率、持续时间和滑动)上有变化。尽管自20世纪60年代以来就观察到了表面蠕变，但产生蠕变的物理过程尚不清楚。已经提出了许多蠕变模型，包括粘性流变学模型7、速率和状态摩擦模型8、尺寸限制摩擦模型9和剪切诱导剪胀模型10。每种模型都有自己的优点。然而，关于哪一种过程产生表面蠕变的争论仍在继续。另一个有待回答的基本问题是，蠕变事件中涉及了多少断层。由于蠕变事件是在点位置用蠕变仪记录的，我们不知道在这些事件期间有多少断层是蠕动的。通过使用互补的蠕变仪，观察到了沿着卡拉维拉斯断层的蠕变传播11。通过研究蠕变传播，这些蠕变事件的大小可能被揭示出来，从而对地表蠕变在更广泛的地震周期中所起的作用有了新的理解。目的确定蠕变事件的大小。区分已提出的蠕变事件的物理模型。实施并发展一种基于时移成像的地表蠕变监测技术。

Faults have been knownto slip aseismicallyor creep at the surface since the 1960s when creep was observed along the San Andreas Fault following the 1966 Parkfield Earthquake 1. The San Andreas Fault has since been observed to be creeping along a 150km long sectionbetween San Juan Bautista and Parkfield. Since the discovery of creep along this section of the San Andreas Fault, creepmeters and alignment arrays have been installed to monitor the creep2. Surface creep has been observed on many other fault systems around the world. Creephas been observed on other large strike-slipfaults such as the North Anatolian Fault near Ismetpasa, Turkey 3and the Philippine Fault on the Island of Leyte 4. Strike-slip faults are not the only type of faults that have surface creep, with creep observed on reverse faults in Taiwan 5,6.Creep on faults has been observed to occur in bursts known as creep events; these can be identified on the recordings oncreepmeters(Figure 1).Creep events have variations in slip characteristics (slip rate, duration,and slip) as a result ofphysical processesthat causes creep events. Despite the observation of surface creep since the 1960s,thephysical processby which creep is produced is not yet known. Many models have been proposed for creep including,viscous rheology 7, rate-and state-friction 8, size-limited friction 9and shear-induced dilatancy 10. Each modelhas its own merits.However,the debate over which one process produces surface creep is still ongoing. Another fundamental question still tobeanswered is howmuch of the fault is involved in creep events are.Given creep eventsare recorded at point locations using creepmeters, we do not know how much of the fault is creeping during these events.By using complementary creepmeters, creep propagation along the Calaveras Faulthas been observed 11. By investigating creep propagation, thesize extent of these creep events maybe uncovered leading to a new understanding of the role surface creep plays in the wider earthquake cycle. AimsDetermine how big creep eventsare.Discriminate between physical models proposed for creep events.Implementand developa time-lapse imaging-based technique for monitoring surface creep.