Earth’s crust typically moves a few millimeters to centimeters per year. In an actively deforming continental region, the crust often behaves like a set of nearly-rigid blocks separated by faults.
After a large earthquake, the crust does not stop moving. The slip that occurs during the aftershocks that follow is called afterslip.
Overview | Tracking Stress Buildup and Crustal Deformation | Fault Slip Rates and Post-Earthquake Motions
Ground Movement and Ground Shaking | Cone Penetration Testing (CPT) | Rock Physics Lab
San Francisco Bay Area Arrays and East Bay Seismic Experiment | Neogene Deformation
Fault Slip Rates
Earth’s crust typically moves a few millimeters to centimeters per year. In an actively deforming continental region, the crust often behaves like a set of nearly-rigid blocks separated by faults. The velocity within a block is nearly constant, but the velocity may differ substantially from one block to an adjacent one. Whether the transition in velocity across a bounding fault is abrupt or gradual depends on whether the fault is creeping (gradual transition) or locked (abrupt transition). The distinction is important because a gradual transition implies not only that the fault is locked but that it is accumulating stress that will eventually lead to earthquakes.
An example is the horizontal crustal deformation field in the northern San Francisco Bay area. The crust is approximated as three blocks, their boundaries lying along the right-lateral strike-slip San Andreas fault, Macaama fault, and Bartlett Springs fault. Motions across the San Andreas fault, southern Macaama fault, and southern Bartlett Springs fault are gradual, but those across the northern Macaama fault and northern Bartlett Springs fault are abrupt in several areas, implying some degree of fault creep.
A recent study by Murray, Minson, and Svarc employs use a Bayesian technique to interpret the GPS crustal velocity field to estimate deep slip rates and shallow surface creep. Their model indicates a deep slip rate of 20 mm/yr for the San Andreas fault, a deep slip rate of 13 mm/yr and shallow creep rate of 0 to 13 mm/yr on the Maacama fault, and a deep slip rate of 7 mm/yr and shallow creep rate of 0 to 7 mm/yr along the Bartlett Springs fault.
Post-Earthquake Motions
After a large earthquake, the crust does not stop moving. Earthquakes such as the 1994 M6.7 Northridge, California, or 1989 M6.9 Loma Prieta, California, earthquakes are followed by hundreds of aftershocks, some of them damaging. Many aftershocks occur on the causative fault or its extension into the deeper earth. The slip associated with these aftershocks is termed afterslip. It may occur in the form of ordinary earthquakes or as slow slip that is not felt.
Another process that occurs after an earthquake is relaxation of the ductile rock that underlies the uppermost crust. While most earthquakes occur in the uppermost crust because frictional resistance is high (and hence stress builds up over long periods of time), few occur in the deeper crust or underlying mantle because the rock temperature is too high to permit accumulation of stress. Instead, stress is accommodated by flow of the rock. The depth where the transition from brittle friction to ductile flow is called the brittle-ductile transition, and it generally coincides with the maximum depth of earthquakes, typically 10-20 km (Implications of the Depth of Seismicity for the Rupture Extent of Future Earthquakes in the San Francisco Bay Area).
Understanding these processes is important for estimating how faults interact with one another through stress transfer, which can change the seismic hazard for years after an earthquake. Part of this understanding comes from better knowledge of the temperature and composition of rocks deep beneath earth’s surface.
Case Study: M7.1 October 16, 1999 Hector Mine, California, Earthquake
The Hector Mine earthquake occurred in the central Mojave Desert and produced several meters of slip along faults of total length ~40 km. It was well-recorded by continuous GPS sites of the SCIGN Network that were already in place in the aftermath of the 1992 M7.3 Landers earthquake, as well as new continuous and survey-mode GPS that were installed in the region by scientists soon after the Hector Mine quake.
Both afterslip on thin zones below the causative fault and relaxation of the flowing lower crust and mantle occurred in the months and years after the Hector Mine earthquake.
2.5 years after the 1999 Hector Mine quake quake, almost all regional motions above the ‘background’ tectonic motions are due to flow of rock in the lower crust and upper mantle. This motion can be tracked unambiguously up to 10.46 years after the 1999 Hector Mine quake, at which time the M7.2 April 4, 2010 El Mayor-Cucapah earthquake changed the regional flow pattern. A model of the flow based on the laboratory behavior of wet olivine and mafic granulite compositions (Thatcher and Pollitz, 2008) agrees well with observed motions.
- Overview
Earth’s crust typically moves a few millimeters to centimeters per year. In an actively deforming continental region, the crust often behaves like a set of nearly-rigid blocks separated by faults.
After a large earthquake, the crust does not stop moving. The slip that occurs during the aftershocks that follow is called afterslip.
Overview | Tracking Stress Buildup and Crustal Deformation | Fault Slip Rates and Post-Earthquake Motions
Ground Movement and Ground Shaking | Cone Penetration Testing (CPT) | Rock Physics Lab
San Francisco Bay Area Arrays and East Bay Seismic Experiment | Neogene Deformation
Fault Slip Rates
Sources/Usage: Public Domain.The USGS has been monitoring the Bartlett Springs fault and Maacama fault with continuous and campaign GPS instruments since 2006. GPS and alinement arrays around these faults are designed to capture long-term movements as well as shallow creep rates. (Public domain.) Earth’s crust typically moves a few millimeters to centimeters per year. In an actively deforming continental region, the crust often behaves like a set of nearly-rigid blocks separated by faults. The velocity within a block is nearly constant, but the velocity may differ substantially from one block to an adjacent one. Whether the transition in velocity across a bounding fault is abrupt or gradual depends on whether the fault is creeping (gradual transition) or locked (abrupt transition). The distinction is important because a gradual transition implies not only that the fault is locked but that it is accumulating stress that will eventually lead to earthquakes.
An example is the horizontal crustal deformation field in the northern San Francisco Bay area. The crust is approximated as three blocks, their boundaries lying along the right-lateral strike-slip San Andreas fault, Macaama fault, and Bartlett Springs fault. Motions across the San Andreas fault, southern Macaama fault, and southern Bartlett Springs fault are gradual, but those across the northern Macaama fault and northern Bartlett Springs fault are abrupt in several areas, implying some degree of fault creep.
A recent study by Murray, Minson, and Svarc employs use a Bayesian technique to interpret the GPS crustal velocity field to estimate deep slip rates and shallow surface creep. Their model indicates a deep slip rate of 20 mm/yr for the San Andreas fault, a deep slip rate of 13 mm/yr and shallow creep rate of 0 to 13 mm/yr on the Maacama fault, and a deep slip rate of 7 mm/yr and shallow creep rate of 0 to 7 mm/yr along the Bartlett Springs fault.
Sources/Usage: Public Domain.Horizontal velocity field in the northern San Francisco Bay area as measured by a combination of continuous (CGPS) and survey-mode (SGPS) GPS. (Public domain.)
Sources/Usage: Public Domain.Shallow and deep fault slip rates inferred for the San Andreas fault, Macaama fault, and Bartlett Springs fault. Only the deep slip rate is shown for the SAF. The long-term slip rate of each fault is ideally its deep slip rate; the shallow portion of each faults catches up to the (generally higher) velocity of the deeper portion episodically, i.e. during earthquakes. The higher the shallow slip rate, the more the fault is creeping, hence the lower the expected size of earthquakes. (Public domain.) Post-Earthquake Motions
Sources/Usage: Public Domain.Left: Afterslip. Right: Lower crust and upper mantle relaxation. Areas colored yellow either slip or flow after a large earthquake. Both the afterslip and relaxation processes occur deep under the faults (schematic red lines) that originally slipped in the earthquake. UC, Upper Crust; LC, Lower Crust; UM, Upper Mantle. The UC/LC boundary is schematically the brittle-ductile transition. Determining whether afterslip or relaxation occur after an earthquake is an active research area. (Public domain.) After a large earthquake, the crust does not stop moving. Earthquakes such as the 1994 M6.7 Northridge, California, or 1989 M6.9 Loma Prieta, California, earthquakes are followed by hundreds of aftershocks, some of them damaging. Many aftershocks occur on the causative fault or its extension into the deeper earth. The slip associated with these aftershocks is termed afterslip. It may occur in the form of ordinary earthquakes or as slow slip that is not felt.
Another process that occurs after an earthquake is relaxation of the ductile rock that underlies the uppermost crust. While most earthquakes occur in the uppermost crust because frictional resistance is high (and hence stress builds up over long periods of time), few occur in the deeper crust or underlying mantle because the rock temperature is too high to permit accumulation of stress. Instead, stress is accommodated by flow of the rock. The depth where the transition from brittle friction to ductile flow is called the brittle-ductile transition, and it generally coincides with the maximum depth of earthquakes, typically 10-20 km (Implications of the Depth of Seismicity for the Rupture Extent of Future Earthquakes in the San Francisco Bay Area).
Understanding these processes is important for estimating how faults interact with one another through stress transfer, which can change the seismic hazard for years after an earthquake. Part of this understanding comes from better knowledge of the temperature and composition of rocks deep beneath earth’s surface.
Sources/Usage: Public Domain.An example of the postseismic motions occurring over the year following a thrust (Northridge-type) earthquake is illustrated here. The co-seismic motions occur suddenly at the time of the earthquake, the postseismic motions gradually after the earthquake through the relaxation process. The earthquake originally occurs in the upper gray layer (low-temperature rock), overlying a deeper high-temperature region. (Public domain.) Case Study: M7.1 October 16, 1999 Hector Mine, California, Earthquake
The Hector Mine earthquake occurred in the central Mojave Desert and produced several meters of slip along faults of total length ~40 km. It was well-recorded by continuous GPS sites of the SCIGN Network that were already in place in the aftermath of the 1992 M7.3 Landers earthquake, as well as new continuous and survey-mode GPS that were installed in the region by scientists soon after the Hector Mine quake.
Both afterslip on thin zones below the causative fault and relaxation of the flowing lower crust and mantle occurred in the months and years after the Hector Mine earthquake.
2.5 years after the 1999 Hector Mine quake quake, almost all regional motions above the ‘background’ tectonic motions are due to flow of rock in the lower crust and upper mantle. This motion can be tracked unambiguously up to 10.46 years after the 1999 Hector Mine quake, at which time the M7.2 April 4, 2010 El Mayor-Cucapah earthquake changed the regional flow pattern. A model of the flow based on the laboratory behavior of wet olivine and mafic granulite compositions (Thatcher and Pollitz, 2008) agrees well with observed motions.
- Publications